a ARAL

Ad Vai ENE A :

Bees

OO

‘+

Sreokeete,

VAT His enunen

LIBRARY OF THE FOR THE PEOPLE

ay <x FOR

2 EDVCATION

%

©

ra : > . » = 2 Send ‘

THIRTY-SIXTH ANNUAL REPORT

OF THE

Mame Aerieultural: Exp ina Mati

ORONO, MAINE

1920 UNIVERSITY OF MAINE

1920

MAINE AGRICULTURAL EXPERIMENT STATION ORONO, MAINE

ORGANIZATION JANUARY To JuNE, 1920 THE STATION COUNCIL

PRESIDENT ROBERT J. ALEY, President DIRECTOR CHARLES. D. WOODS, Secretary ORA GILPATRICK, Houlton, | 3

FRANK E. GUERNSEY, Dover, \ Comariiices os, CHARLES S. BICKFORD, Belfast, \ Board of Trustees JOHN A. ROBERTS, Commissioner of Agriculture EUGENE H. LIBBY, Auburn, State Grange WILSON H. CONANT, Buckfeld, State Pomological Society FRANK S. ADAMS, Bowdoinham, State Dairymen’s Association

LEONARD C. HOLSTON, Cornish, Maine Livestock Breeders’ Assn. WILLIAM G. HUNTON, Portland, Maine Seed Improvement Assn.

AND THE HEADS AND ASSOCIATES OF STATION DEPARTMENTS, AND THE ; DEAN OF THE COLLEGE OF AGRICULTURE

THE STATION STABE

CHARLES D. WOODS, Sc D. Director

ADMINIS- ESTELLE M. GOGGIN, Clerk TRATION ‘) CHARLES C. INMAN, Clerk | MARY L. NORTON, Clerk

JOHN W. GOWEN, Pz. D., Biologist

RAYMOND PEARL, Px. D., Collaborator

BIOLOGY % wiILDRED R. COVELL, Clerk HELEN A. RING, Laboratory Assistant

JAMES M. BARTLETT, M. S.,, Chemist

CHEMISTRY EDM RS Rei OR ED Yaa aaS= Assistant (@ISVAIRP AN WABI Pier. (Ce Assistant

ENTOMOL- ( EDITH M, PATCH, Pu. D. Entomologist OGY | ALICE W. AVERILL, Laboratory Assistant

su ( WARNER J. MORSE, Pu. D, Pathologist ELANT - ) DONALD FOUSOM Pram Assistant FATHOLOGY | SOLA, MORRIS: Laboratory Assistant = ( JACOB ZINN, Acer. D., Assistant Biologist AROOSTOOK ) & RAYMOND RING, A. B., Saienie Aid FARM / WALTER E. CURTIS, Superintendent HIGHMOOR ( WELLINGTON SINCLAIR, : Superintendent FARM ) HUGH C. McPHEE, B. S., Scientific Aid

ROYDON L. HAMMOND, Seed Analyst and Photographer

MAINE AGRICULTURAL EXPERIMENT STATION . ORONO, MAINE

ORGANIZATION JULY To DecemBer, 1920 THE STATION COUNCIL

PRESIDENT ROBERT J. ALEY, erecident DIRECTOR CHARLES D. WOODs, Succ ORA GILPATRICK, Houlton,

CHARLES FE. OAK, Bangor, Committee of

THOMAS E. HOUGHTON, Fort Fairfield, Board of Trustees

Commissioner of Agriculture

eoeeee ese eo ee eo ee ee ee ee ew

EUGENE H. LIBBY, Auburn, State Grange - WILSON W. CONANT, Buckfield, State Pomological Society FRANK S. ADAMS, Bowdoinham, State Dairymen’s Association

LEONARD C. HOLSTON, Cornish, Maine Livestock Breeders’ Ass'n. WILLIAM G. HUNTON, Portland, Maine Seed Improvement Ass’n.

Anp THE HeEAps AND ASSOCIATES OF STATION DEPARTMENTS, AND THE DEAN OF THE COLLEGE OF AGRICULTURE

THE STATION STAFF

CHARLES D. WOODS, Sc D. Director

ADMINIS- ESTELLE M. GOGGIN, Clerk TRATION CHARLES C. INMAN, Clerk MARY L. NORTON, Clerk

JOHN W. GOWEN, Pu. D., Biologist

KARL SAX, M. S., Biologist

BIOLOGY % MILDRED R. COVELL, Clerk BEATRICE GOODINE, - Laboratory Assistant

JAMES M. BARTLETT, M. S., Chemist

CHEMISTRY ELMER R. TOBEY, Cu. E., Associate C. HARRY WHITE, Pu. C., Assistant

ENTOMOL- EDITH M. PATCH, Pu. D., Entomologist OGY ALICE W. AVERILL, Laboratory Assistant PLANE WARNER J. MORSE, Pu. D., Pathologist PATHOLOGY DONALD FOLSOM, Pu. D., Associate VIOLA L. MORRIS, Laboratory Assistant

AROOSTOOK JACOB ZINN, Aacr. D., Associate Biologist FARM E. RAYMOND RING, A. B., Superintendent HIGHMOOR WELLINGTON SINCLAIR, Superintendent FARM HUGH C. McPHEE, B. S., Scientific Aid

ROYDON L. HAMMOND, Seed Analyst and Photographer

The publications of this Station will be sent free to any address in Maine. All requests should be sent to Agricultural Experiment Station, Orono, Maine.

CONTENTS. Oraaimizawioal Cir We Sicwioringgacocecdomesooscepadod odo cece daccos Mi OUMCEMIGND 65 od Gute ob Oe OOo Oe ae eee Ai aoe in Tans Ges ree EMDlicakionsmssuedsdralO20)ceeynr vis ceibe te van aclace ame incucromee eal SEBO. INOS S'S AG es A rie Oe cede cy se re A a pe

Wheat Investigations. I. Pure Lines. Summary. (Bulletin 285).. Glineterand) soil selations (Bulletin 285)).5.-45-s20..5-5002- 220-5 Characteristics of Aroostook wheats (Bulletin 285).............. Origin of pure lines at Aroostook Farm (Bulletin 285)............ Analysis ot Cea Cexnillenim AS)\>psoncsocdssoscousoceusccvceeonoues @hemical characters, or pure) lines (Bulletin 285) ......75200.-50--- Bakancarestsmorethespune) limes! (BulletingZ85) 44 45erees aan ace Relation between protein, gluten content and size of loaf

CB eaillligraira: Aoi) ee ancl tenc votre each otc ee OLSEN te sl are ton ee ear Ce ate Discissronmandaconclusions. (Bulletin 285)ieqnee soe eee see The Variation of Milk Secreticn with Age in Jersey Cattle.

Sumamevay,: 2 (CBiisate ;AN0)) 56 oe a6 sow oo boo duoa nmonadEoheuooe oT Varmationmat commencement of test.) (Bulletin: 286s. 2-96 >5.- 40. Correlation of 8 months’ milk production with age at test.

GB lle tine eZS6) yeaa eee eevee aie ee tela te Mae ieee ek Regression of milk production on age (Bulletin 286) SOR aye Le Self Sterility and Cross Sterility in the Apple... Summary.

(Byaillkeratin: “ZAY) a real ee crete ray els tv ee Oy ee eR Ta Self sterility and self fertility (Bulletin DBT 2) tere Wert cays ae NRE aR Ac Grossmsteniityerand across. fertility: (Bullletin= 287) 2 1. ao Vigor and size of apples from selfed or crossed varieties

Bulletin Z SI) vercncrice eae eater ie ses ee cee suai UR Sar ese) Causes of self sterility and cross sterility (Bulletin 287)............ Some Observations Upon the Effect of Borax in Fertilizers.

Simmmanrn “Cpnilleiim: AIS) sosuctadocobpenebeenoobeeoceueuoones lmmoctneniom CB Ewieaia sA7ets))n 0 hate ws soe balan bonis HedE ea ae oiomame cs Field observations on potatoes in 1919 (Bulletin 288).............. Injury in relation to fertilizer used (Bulletin 288).................. Greenhouse experiments with fertilizers containing borax

GBilletinipZ88) es sees seacaset acre persiseeto manic eas cea tee ed Soe ROR AG eats A Voramvitoepotatoesm Bulletin Z288)\neene eens eens ania Results from greenhouse experiments with potatoes (Bulletin 288) Work wath Odier Cros (Cswiilesin AS) ccosnonescsashecoeschoosuc Beams CRnillesim Ayo cesceccoocebsnscosndsoe eae Sa AI eed a eee Sate Sit Ortcwvheatandubuckwheate (Si ulletineZS8) pases eee aoe eee The Correlation Between Milk Yield of One Lactation and That

of Succeeding Lactations. Summary. (Bulletin 289).......... Correlation of milk production of one age with any other age

GBiilletinte ZS 9 ranma cere cutee Cyaan eeear MLN Irae MMMM Nes aa

Correlation of one lactation record with the milk yield for the first

five. lactations: ((Bulletin] 289). 2. ee eee eee 129 ‘The Variation of Butter-Fat Percentage w ith Age in Jersey Cattle

Summary o(Bulletin2290)).. 2 be ne eee 132 iNatenials cand methods) (Bualletine 290) 295.5 eee ee 135 ‘Variations of fat percentage with the age when test was made

€Balletin’ 290). soccc be. Seas seen eee ee ee ee eee 136 Correlation of 8 months’ butter-fat per cent with age cf cow

(Bulletin 290) Say crosses dig tec needle se he CEE 142 The Correlation Between the Butter-Fat Percentage of One Lacta-

tion and Succeeding Lactations in Jersey Cattle. Summary.

(Bulletin 290) 52. ss wale oo ee Eee 145 Correlation of butter-fat per cent for 8 months with like butter-fat

per cent at any, other sivemage. (Bulleting291) as ae 147 Correlation between butter-fat per cent cf first five lactations and ~

the mean butter-fat per cent of these individual lactations

(Bulletin 291). 2 s.. cc 22k pels s een ee ee 152 Potato- Mosaic, “Summary. 9 (Bulletin 292),..22 25s eee eee 157 Introduction: CBullétin. 292) ih: 2 52a eee 158 Appearance of the diseased plants (Bulletin 292)...... tea ae 160 @ther jeffects of the disease ((BulletinZ292) he acer eee 161 Lransmission ibyathe tubers, ((BulletinmZ97) ease eee ce eee eee eoee 163 Proots of mrectiousness (Bulletin 292) pees eee eae eee eee 164 Insects as) carniers. (Bulletin, 292) >.<. sone eee Coe eee eee eee 168 Other possible factcrs in the spread of mosaic (Bulletin 292)...... 175 Methods Gf ‘control, (¢Bulletin, 292)- 22a eee eee _ 180

Recommendations for the control of potato mosaic (Bulletin 292) 183 Studies in Milk Secretion VIII. Influence of Age on Milk and

Butter-Fat Yield in Helstein-Friesian Cattle (Bulletin 293)... 185 Normal and abnormal germination of grass-fruits. Summary. (Bulletin: 294) 3c. hae ES ee ee eee 197 Material and) Methods (Bulletin 294) px 2ee eee ee eee 199 Normal Germination of Grass-Fruits (Bulletin 294).............. 199 Abnormal Germinaticn of Grass-Fruits (Bulletin 294)............ 207 The Mechanical and Biological Functions of the Coleorhiza i (Bulletin” 294) oo hen. Ee ba a ee 213 Polyemryony in ‘Grasses - (Bulletin) 294) 22a eee eee 214 Abnormal Germination as a Source of Error in Germination Tests (Bulletin: 294). 250.0 8 oa ee Ee eee eee 214

Transmission cf milk yield to the fees generation (Bulletin 295).. 217 Transmission of butter-fat percentage in the first generation

(Bulletin 295). os jocesc ce w le sere cleo sete SO RRO ee ee eee 219 Mode of Transmission of milk quantity as shown by first genera-

ation! (Bulletin 295); scc26e0 <a<clon- cake Ree EE Coen 221 The life cycle of Aphids and Coccids (Bulletin 295).............. 221 ‘Transmission of the Mosaic disease of Irish potatces (Bulletin 295) 223 Meteorological observations (Bulletin 295).....c...........-.... 224 iReport of Treasurera( Bulletin’ 295), 250-2 ¢- eee oe eee eee eee 226

Gnadexss 7 CBulletim wZ95)).c.. osc. Cee eee eee 229

f

ANNOUNCEMENTS.

ESTABLISHMENT OF THE STATION

“The: Maine Fertilizer Control and Agricultural Experiment Station, established by Act of the Legislature approved March 3, 1885, began its work in April of that year in quarters fur- nished by the College. After the Station had existed for two

" years, Congress passed what is known as the Hatch Act, estab-

lishing agricultural experiment stations in every state. This grant was accepted by the Maine Legislature by an Act ap- proved March 16, 1887, which established the Maine Agricul- tural Experiment Station as a department of the University. The reorganization was effected in June, 1887, but work was not begun until February 16, 1888. In 1906, Congress passed the Adams Act for the further endowment of the stations es- tablished under the Hatch Act.

The purpose of the experiment stations is defined in the Act of Congress establishing them as follows:

“Tt shall be the object and duty of said experiment stations to conduct original researches or verify experiments on the phy- siology of plants and animals; the diseases to which they are severally subject, with the remedies for the same; the chemical composition of useful plants at their different stages of growth; the comparative advantage of rotative cropping as pursued un- der a varying series of crops; the capacity of new plants or trees for acclimation; the analysis of soils and water; the chemical composition of manure, natural and artificial, with experiments designed to test their comparative effects on crops of different kinds; the adaptation and value of grasses and forage plants; the composition and digestibility of the different kinds of food for domestic animals; the scientific and economic questions in- volved in the production of butter and cheese; and such other researches or experiments bearing directly on the agricultural industry of the United States as may in each case be deemed advisable, having due regard to the varying conditions and needs of the respective states or territories.”

Vili MAINE AGRICULTURAL EXPFRIMENT STATION.

The work that the Experiment Station can undertake from the Adams Act fund is more restricted and can “be applied only to paying the necessary expenses for conducting original researches or experiments bearing directly on the agricultural industry of the United States, having due regard to the vary- ing conditions and needs of the respective states and territories.”

INVESTIGATIONS.

The Station continues to restrict its work to a few impor- tant lines, believing that it is better for the agriculture of the State to study thoroughly a few problems than to spread over the whole field of agricultural science. It has continued to im- prove its facilities and segregate its work in such a way as to make it an effective agency for research in agriculture. Promi- nent among the lines of investigation are studies upon the food of man and animals, the diseases of plants and animals, breed- ing of plants and animals, orchard and field experiments, poul- try investigations, and entomological research.

INSPECTIONS.

Up to the close of the year 1913, it had been the duty of the Director of the Station to execute the laws regulating the sale of agricultural seeds, apples, commercial feeding stuffs, commercial fertilizers, drugs, foods, fungicides and insecticides, and the testing of the graduated glassware used by creameries. Beginning with January, 1914, the purely executive part of these laws is handled by the Commissioner of Agriculture. It is still the duty of the Director of the Station to make the an- alytical examination of the samples collected by the Commis- sioner and to publish the results of the analyses. The cost of the inspections is borne by fees and by a State appropriation.

OFFICES AND LABORATORIES.

The offices, laboratories and poultry plant of the Maine Agricultural Experiment Station are at the University of Maine, Orono. Orono is the freight, express, post, telegraph and tele- phone address for the offices and laboratories.

ANNOUNCEMENTS. 1X

AROOSTOOK FARM.

- By action of the Legislatures of 1913 and 1915 a farm was purchased in Aroostook County for scientific investigations in agriculture to be under “the general supervision, management, and control” of the Maine Agricultural Experiment Station. The farm is in the town of Presque Isle, about 2 miles south of the village, on the main road to Houlton. The Bangor. and Aroostook railroad crosses the farm. A flag station, “Aroos- took Farm,” makes it easily accessible by rail.

The farm contains about 275 acres, about half of which is cleared. The eight room house provides an office, and home for the farm superintendent. A school house on a lot adjoining the farm was presented to the State by the town of Presque Isle and after being remodeled serves as a boarding house for the help. A greenhouse and a potato storage house have been erected at the farm by the U. S. Department of Agriculture for use in cooperative work on potato breeding. The large barn affords storage for hay and grain and has a large potato storage house in the basement.

HicHMmoor Farm.

The State Legislature of 1909 purchased a farm upon which the Maine Agricultural Experiment Station was directed to “conduct scientific investigations in orcharding, corn and other farm crops.” The farm is situated largely in the town of Mon- mouth. It is on the Farmington Branch of the Maine Central Railroad, 2 miles from Leeds Junction. A flag station, “High- moor,” is on the farm. :

The farm contains 225 acres, about 200 of which are in orchards, fields, and pastures. There are in the neighborhood of 3,000 apple trees upon the place which have been set from 20 to 30 years. The house has 2 stories with a large wing, and con- tains about 15 rooms. It is well arranged for the Station offices and for the home of the farm superintendent. A substantially ‘constructed building for apple packing was erected in 1912.

The removal of the crossbred herd from the University to _ Highmoor necessitated considerable change in the barns and the building of a new one 80 x 36 to accommodate the herd. This

5:< MaINne AGRICULTURAL EXPERIMENT STATION.

barn has a basement for manure, the cow stanchions above, and a loft for storage of hay. The silo has been enlarged and a long shed has been made into calf pens. A well to supply the necessary water has been driven.

PUBLICATIONS.

The Station is organized so that the work of investigation is distinct from the work of inspection. The results of investi- gation are published in the bulletins of the Station and in sci- entific journals, both foreign and domestic. The bulletins for the year make up the annual report. The results of the work of inspection are printed in publications known as Official In- spections. These are paged independently of the bulletins and are bound in with the annual report as an appendix thereto. Miscellaneous publications consisting of newspaper notices of bulletins, newspaper bulletins and circulars which are not paged consecutively and for the most part are not included in the an- nual report are issued during the year. Weekly mimeograph publicity letters are sent to all papers within the State.

BULLETINS ISSUED IN 1920.

No. 285. Wheat Investigations. I. Pure Lines. 48 pages. 3 pages of plates.

No.’ 286. The Variation of Milk Secretion with Age in Jersey Cattle. 9 pages.

No. 287. Self Sterility and Cross Sterility in the Apple. 20 pages.

No. 288. Some Observations Upon the Effect of Borax in Fertilizers. 33 pages.

No. 289. The Correlation Between Milk Yield of One Lactation and That of Succeeding Lactations. 10 pages.

No. 290. The Variation of Butter-Fat Percentage with Age in Jersey Cattle. 12 pages.

No. 291. The Correlation Between the Butter-Fat Percentage of One Lactation and Succeeding Lactations in Jersey Cattle. 9 pages.

No. 292. Potato Mosaic. 28 pages.

No. 293. Studies in Milk Secretion VIII. Influence of Age on Milk and Butter-Fat Yield in Holstein-Friesian Cattle. 13 pages.

No. 294. Normal and Abnormal Germination of Grass-Fruits. 19 pages. 4 pages of plates.

PUBLICATIONS. exh

No. 295. Abstracts of Papers not included in Bulletins, Finances, Meteor-- ology, Index.

OFFICIAL INSPECTIONS ISSUED IN 1920.

No. 95. Drugs and Foods. 28 pages.

No. 96: Commercial Feeding Stuffs, 1919-20. 37 pages. No. 97. Commercial Fertilizers, 1920. 25 pages.

No. 98. Commercial Agricultural Seeds, 1920. 25 pages.

MISCELLANEOUS PUBLICATIONS ISSUED IN 1920.

No. 538. The Relation of Conformation to Milk Yield in Jersey Cattle.. ; 12 pages. No. 539. Improved Strains of Aroostook Grown Wheats. 11 pages.

BIOLOGICAL PUBLICATIONS, 1920.

In the numbered series of “Papers from the Biological Laboratory”:

132. Wheat Investigations. I. Pure Lines. By Jacob Zinn. Annual: Report of the Maine Agricultural Experiment Station for 1920. Bulletin 285, pp. 1-49.

133. Self Sterility and Cross Sterility in the Apple. By John W. Gowen. Annual Report of the Maine Agricultural Experiment Station for 1920. Bulletin 287, pp. 61-89.

134. Studies in Milk Secreticn VIII. On the Influence of Age on Milk Yield and Butter-Fat Percentage, as Determined from the 365. day Records of Holstein-Friesian Cattle. By John W. Gowen. Annual Report of the Maine Agricultural Experiment Station: for 1920. Bulletin 293, pp. 185-196.

135. Inheritance in Crosses of Dairy and Beef Breeds of Cattle. II. On the Transmission of Milk Yield to the First Generation. By- John W. Gowen. Journal of Heredity.

136. Inheritance in Crosses of Dairy and Beef Breeds. III. Trans- mission of Butter-Fat Percentage to the First Generation. By John W. Gowen. Journal of Heredity.

ENTOMOLOGICAL PAPERS, 1920.

Ent. No. 106. The Life Cycle of Aphids and Coccids. By Edith M. ; Patch. Annals Entomological Society cf America, Vol.. 13, No. 2. pages 156-167.

STATION NOTES. COUNCIL AND STAFF CFEANGES.

At the June meeting of the Trustees Mr. Chas. E. Oak and’ Mr. Thomas E. Houghton were appointed to represent the

xii Matne AGRICULTURAL EXPERIMENT STATION.

Board of Trustees on the Station Council in place of Frank E. Guernsey and Chas. S. Bickford.

Roydon L. Hammond, Seed Analyst and Photographer, re- signed on October I, to accept a similar position with the Dela- ware State Department of Agriculture.

Dr. Raymond Pearl, Collaborating Biologist for the Station, resigned this position in June. The poultry records and much other material which he was preparing for publication in bulle- tins of the Maine Station were all lost in a fire which destroyed the old group of buildings at Johns Hopkins University early in the year.

On December Ist, after nearly twenty-five years of efficient service as Director of the Maine Experiment Station, Dr. Chas. D. Woods’ term of office was abruptly terminated. This unfor- tunate occurrence came at a very inopportune time. The year’s work had been nearly completed but the Annual Report which was always arranged by him had not been prepared. The Re- port consists largely of bulletins issued at intervals through the year giving the results of most of the work done by the Station. The Director, however, carried on some experimental work un- der his own supervision the results of which no one else was very familiar. This had not been prepared for publication and is consequently omitted from the Report.

Soit Test EXPERIMENT.

The soil test experiment at Aroostook Farm which was started in 1916 is still in progress. A new piece of land,.how- ever, has been taken for the purpose. The first piece selected gave such uneven yields that we were forced to the conclusion that the soil lacked the necessary uniformity for this type of work. The new piece was divided into plots and planted to potatoes in 1919, without fertilizer, to test its uniformity. The yields (on the different plots) were quite uneven but whether these differences were due to previous treatments and will dis- appear with further cropping cannot now be told. In 1920 all the plots were dressed with ground limestone, sowed to oats and seeded down to clover. The yields of oats were somewhat more even than those of the potatoes but not so eyen as desirable and further cropping may be necessary before the experiment with chemicals can be begun.

BULLETIN 285

WHEAT INVESTIGATIONS. I. PURE LINES.’

By Jacos ZINN.

SUMMARY

The present bulletin contains an account of the origin and development of a number of pure lines of wheat by the method of selection, including data on all the important stages, except milling, from the single head selection to the bakehouse. The study of the effect of the environment of Northern Maine upon the physical and chemical characteristics of pure strains intro- - duced from Minnesota forms an incidental feature of this: report.

In 1915 several hundred wheat spikes were selected from. commercial varieties representing the chief groups of hard red. spring wheat. Of these selections 259 heads were retained and. each grown in a row in the cereal crop nursery in 1916. In the following year 91 strains of the original selections were grown. in one two-thousandth acre plots, along with 7 pure strains introduced from Minnesota. The crop from the one two- thousandth acre plots furnished enough seed to make a chemi- cal determination of the crude protein content of each strain. In 1918 only 44 pure lines selected from Aroostook wheats and 6 lines of Minnesota wheats were retained and grown in plots ranging from one two-hundredth to one-fortieth acre in area. A complete chemical analysis of the wheat and flour of 37 lines and a baking test of 31 lines were made in the spring of 1919..

Under the same conditions of environment the pure lines of wheat showed distinct differences in the physical and chemi- cal characteristics and in the bread making value of their grain.

The average weight per 1000 kernels for all lines was found. to be 35.314 grams. The individual strains within a variety: showed a very considerable variation in the weight per 1000:

‘Papers from the Biological Laboratory of the Maine- Agricultural: Experiment Station, No. 132.

2 Marne AGRICULTURAL EXPERIMENT STATION. 1920

kernels—ranging from 26.541 to 44.789 grams—as well as a marked deviation from the average of their respective parent varieties. The strains with the highest weight per 1000 kernels produced the greatest percentage of yellow berries and yielded flours of poor baking quality. The environmental conditions prevailing in Aroostook brought the low kernel weight of the original Minnesota seed up to the level of the Aroostook strains within a single season. This change, however, was not found to be progressive.

The data on the yield, though very limited, show a number of strains of high yielding capacity. Each variety furnished high and low yielding strains the differences in yield between the lines of the same variety being greater than between varie- ties. .

The average protein content of the Aroostook lines was 13.81 and 12.62 per cent for the season of 1917 and 1918, re- spectively. The Canada Red (Ladoga type) and Preston strains yielded the highest, the Marquis the lowest protein content.

A study of the relationship between the protein content of the pure lines in 1917 and 1918 revealed a tendency for the varieties as well as for the strains to retain their relative rank with respect to protein content from one year to the next. The coefficient of correlation between the protein content of the pure strains in 1917 and 1918 was found to be 0.381.092.

Certain strains of bread wheats introduced from Minne- sota retained their high protein content under Aroostook con- ditions. The average protein content of the Aroostook grown Minnesota bread wheats was somewhat higher than that of the Aroostook pure lines for two successive seasons. The durum strains, however, showed a very rapid deterioration. The Min- nesota grown durum strain, Speltz Marz, headed the list of the Minnesota introductions in regard to protein content; at the end of a single season’s growth in Aroostook this line showed the second lowest protein content of all 99 strains analysed. The second durum strain, Hedge Row, showed the lowest protein content of all lines at the end of the first season under Aroos- toook conditions. The low protein content was accompanied by the production of a very high percentage of yellow berries.

A number of the pure lines showed a high gluten content, the Preston strains ranking highest being followed by the Min- nesota, Red Fife, Canada Red, Bluestem, and Marquis lines in

q

WHEAT INVESTIGATIONS. 3

the order named. The Red Fife and Bluestem furnished a number of strains with good quality glutens while the Preston and Marquis yielded a large percentage of strains with fair to poor quality of gluten. The Minnesota lines, with the excep- tion of the durum and Marquis wheat yielded a strong elastic gluten of good quality.

The baking test showed very marked variations in the flour strength of the different pure lines, the volume of bread loaf, baked from 340 grams of flour, ranging from 1518 to 2221 cubic centimeters. A number of strains produced bread possessing a good volume and very good appearance, good to very good _ texture and color of crumb, and excellent eating qualities. Some of the Minnesota strains produced a very good volume and showed very good baking quality. |

The available data indicate that strains of wheat of good quality can be isolated and successfully grown under Aroos- took conditions.

INTRODUCTION.

One of the most perplexing problems connected with agri- cultural crops is the quality of wheat. Though wheat has per- haps a wider range of distribution throughout the world than any other crop, yet the regions in which strong wheats of best quality are successfully grown, are very distinct and rather lim- ited. This relation between the quality of wheat and the en- vironment is of greatest concern primarily to the plant breeder who attempts the improvement of the strength of wheat in re- gions outside the natural districts of the hard red wheats. It is a well known fact that the majority of wheats grown in the forested regions of the northeast are at best semi-hard and do not come up in their bread-making quality to the high standards - of the wheats grown in the prairie regions of the Northwest and Russia. As a direct effect of the influence of the environ- ment upon the wheat grain it has been frequently observed that when some of the hard vitreous wheats from the prairie regions are introduced into localities marked by a cool summer, abun- dant rainfall and high relative moisture, they tend to lose their translucency and horny texture and assume a plump, dull opaque appearance.

4 MaIne AGRICULTURAL EXPERIMENT STATION. 1920

As the baking quality or strength of wheat has been com- monly regarded as being determined by the chemical composi- tion, the problem of strength in wheat has long been studied from the chemical point of view. Two aspects of this problem were especially subjected to a frequent study, viz., the relation between the chemical composition and the bread-making value, and the influence of the environment upon the chemical com- position of wheat. In the first stage of the study of the relation between the chemical composition and strength the quantity of protein and gluten was regarded as the determining factor of strength. As the accumulating evidence on this point was not concordant the investigators in this field turned to the study of the quality of gluten, notably its chemical quality. Various theories have been suggested in explanation of the strength of wheat, such as the gliadin number 1. e. the ratio of gliadin to glutenin, the absolute amount of gliadin in the flour, the ratio of nitrogen to available potassium, the nitrogen in amino form, _ the ratio of total nitrogen to soluble nitrogen in flour, enzymic activity, etc. The study of the physical properties of the gluten gained an impetus since the work of T. B. Wood?, who after establishing that the gliadin and glutenin of strong and weak flours were identical, found that the strength of wheat flour “is associated with a high ratio of proteid to salt and that the size of the loaf depends in the first instance on the amount of sugar contained in the flour together with that formed in the dough by diastatic action.”

In studying the extensive literature on the chemistry of he strength in wheat one must be impressed with the marked diver- gence of views on this problem despite the careful work and standard analytical methods. In view of this the plant breeder is naturally inclined to suspect the heterogeneous nature of the material as being responsible for the conflicting results. For in the study of the chemistry of strength the individuality of the wheat variety that furnished the flour, its inherent specific in- fluence upon the baking value, has been generally neglected. And yet it would seem more reassuring to the plant breeder if in the investigation of such a subtile problem as strength in wheat an inductive rather than deductive procedure be adopted

"Wood, T. B. The Chemistry of Strength of Wheat Flour. Jour- Agric. Sci. 1907 Pt. I., pp. 139-160 and II pp. 267-277.

WHEAT INVESTIGATIONS. 5

by analyzing a given variety of wheat into its component strains and determining the behavior of the flour from each strain under identical conditions. In tracing down the specific behavior of the flour from isolated, individual strains or pure varieties certain varietal features may be established which otherwise may completely elude detection in the analysis of commercial samples of flours furnished by different wheat varieties. What- ever the cause of the discordant results, the fact remains that at the present time there is no commonly accepted, reliable chem- ical formula that could guide the plant breeder in detecting the strength in wheat.

The plant breeder who attempts the improvement of the wheat quality in regions outside the natural wheat lands will find still less encouragement if he turns to results and conclu- sions bearing on the influence of the environment upon the chemical composition of wheat. Here, again, the opinions of the workers are at variance. Without reviewing here the ex- tensive literature relative to the question of climate and soil relations to the chemical composition it may be stated that the majority of investigators, notably Lyon*, Thatcher*, Le Clerc and Leavitt?, Shaw and Walters®, Le Clerc and Yoder’, con- sider the climate as the predominating factor controlling the chemical composition of wheat to the negligence or even exclu- sion of the soil factors. As an interesting event in this connec- tion it may be noted that the agronomic workers who met in the third western agronomic conference at Cornwallis, Oreg.

1918, agreed that quality of wheat was dependent on both soil

and climatic conditions.®

"Lyon, T. L. Improving the Quality of Wheat. U.S.D. Agr. Bur. Plant Ind. 1905 Bull. 78, pp. 1-120.

*Thatcher, R. W. The Chemical Composition of Wheat. Wash. Agr. Pxpe Stay Bully Nos dl. pp. E79) 1913:

"Le Clerc, J. A. and Sherman Leavitt. Tri-local Experiments on the Influence of Environment on the Composition of Wheat, 1910. U.S. Dept. Agr. Bur. Chem. Bull. 128 pp. 1-18.

°Shaw, C. W. and Walters, E. H. A Progress Report Upon Soil and Climatic Factors Influencing the Composition of Wheat. Cal. Agric. Exp. Sta. Bull. 216. pp. 549-574, 1911.

"Le Clerc, J. A. and Yoder, P. A. Environmental Influences on the Physical and Chemical Characteristics of Wheat. 1914. Jour. Agr. Res. V. I, No. 4, pp. 275-291.

‘Jour. Am. Soc. Agr. V. 10, No. 7-8, 1918, p. 312.

—

:

6 Marine AGRICULTURAL EXPERIMENT STATION. 1920

As in the study of the relation of chemical composition to quality so also in most of the investigations into the effect of environment upon chemical composition very little significance was attached to the seed as a factor influencing the composition of the crop. The conclusions reached in these investigations are well reflected in Le Clere’s statement that “soil and seed play a relatively small part in influencing the composition of crops.” (L.c. 1910 p. 18), and that “environment rather than what has been usually termed heredity is the major factor in determining the physical and chemical characteristics of the wheat crop.” (CES, OI, Dp, Zon). |

In view of the generally accepted relation between the chemical composition and quality of wheat these results may become of some concern to the plant breeder. However, in this and similar work again the conclusions. are based on the evi- dence obtained from samples of commercial varieties of wheat. The few experiments with pure strains of wheat entering into this work were concerned with the cumulative effect of selection rather than with the selection and retention of prepotent strains. A common feature of most of the older publications on the in- fluence of environment upon the chemical composition is that they are not accompanied by baking tests, hence do not bear directly on the quality of wheat.

With the advent of the modern principles of plant breeding a third factor, namely, the inherent characteristics of the differ- ent varieties and strains, entered into the consideration of causes influencing the quality of wheat. The first important question confronting the plant breeder was, whether the quality of wheat was merely a function of the environment indiscriminately lev- elling it regardless of the individual characteristics of the differ- ent varieties or strains. Biffen® first subjected the physical char- acters associated with strength to a genetic analysis and found that “strength” and “weakness” form a pair of Mendelian char- acters. Upon these theoretical results a number of hard, cross- bred strains have been built up and tested in the bakehouse. The practical significance of the application of modern breeding principles in the improvement of the quality of wheat is further illustrated by the experiments of the Home Grown Committee

*Biffen, R. H. On the Inheritance of Strength in Wheat. Jour. Agr. Sci. 1908. V. III pp. 86-101.

WHEAT INVESTIGATIONS. 7

of the National Association of British and Irish Millers’? which demonstrated that a number of varieties, notably the Red Fife selections retain their original strength under all conditions when other varieties change enormously with climate and soil. Sim- ilar results: were obtained by Howard, Leake and Howard" in India who found that among 25 pure line cultures representing as many distinct wheats, when grown under different condi- tions at three stations, some strains always remained soft, some had a tendency to remain hard while with the majority of these strains the consistancy varied greatly according to the locality and the conditions under which they are grown (I..c. p. 59). Four strong wheats and one soft strain were each grown at 9 different stations under widely varying conditions of climate, soil and culture. The 4 strains consistently retained their strength and milling qualities while the soft strain always re- mained a weak wheat.

As a further illustration of the effectiveness of the applica- tion of modern plant breeding methods in the improvement of wheat quality the results at the Central Experimental Farm at Ottawa, Ontario, those of Farrer in New South Wales, of Clark '* and of Leith’® should be cited.

The results of a recent investigation of Freeman’ are of special interest in their bearing upon the influence of environ- ment upon the texture of the wheat kernel. In his experiments involving crosses between soft wheats and durum, Freeman es- tablished two types of soft grains. One type was designated by him as “true softness” in which the air spaces in the endosperm are diffuse and finely scattered. This type of softness is only slightly affected by environic conditions. The second type, com- monly called “yellow berry’ was characterized by air spaces

“Humphries, A. E. and Biffen, R. H. The Improvement of English Wheat. Jour. Agr. Sci. 1917, V. 2, pp. 1-16.

: “Howard, Albert, Leake, H. M. and Howard, G. L. C. The Influ- ence of the Environment on the Milling and Baking Qualities of Wheat in India. Memoirs Dept. Agr. India 1913. Vol. V. No. 2, pp. 45-102.

“Clark, J. A. Improvement of Ghirka Spring Wheat in Yield and Quality. 1916. U.S. Dept. Agric. Bu. Plant. Ind. Bull. 450, pp. 1-19.

“Leith, B. D. The milling and Baking Qualities of Wisconsin Grown Wheats. Wisc. Agr. Exp. Sta. Res. Bull. 43, 1919. pp. 1-38.

“Freeman, Geo. F. Producing Bread Making Wheats for Warm Climates, Jour. Heredity, 1918 V. 9, No: 5 pp. 211-226.

0S EE ms mc

8 MatIne AGRICULTURAL EXPERIMENT STATION. 1920

within the endosperm occurring in flakelike groups with quite definite margins, causing a more or less extending opaqueness. This type was found to be very sensitive to environic conditions. Both types were found to exhibit a distinctly different genetic behavior controlled by different sets of genetic factors. The practical importance of these results is apparent for they draw a distinct line between true soft wheats like Sonora, Early Bart which are not affected by climate and are every year 100% soft, and the hard wheats whose response to environmental influences manifests itself in a greater or less percentage of “yellow ber- esha

The present paper deals with the results obtained in the work with a number of pure lines of wheat originated and grown at the Aroostook Farm of the Maine Agricultural Ex- periment Station. The main object of this work was to attempt to improve the strength of the Aroostook wheats, and this bul- letin may be regarded in a way as a progress report on that phase of this work based on the method of pure line selection. In presenting the results of the chemical analyses the writer wishes to emphasize the fact that these data reflecting as they do the chemical composition and behavior of the wheats and lours, all refer to pure strains of wheat in distinction to com- mercial varieties and flours. Some observations on the effect of the environmental conditions of Northern Maine upon a few pure strains of wheat introduced from Minnesota are also here reported.

CLIMATE AND Sort RELATIONS IN Aroostook County.

As already stated the growing of wheat in Maine is con- fined to its northern section made up chiefly of Aroostook ‘County. In view of the exceptional significance commonly at- tached to the environment in relation to the quality of wheat, a brief consideration of the climatic and soil factors of Northern Maine appears desirable.

Northern Maine is characterized by a cool, and moist cli- ‘mate and a short growing season. The mean temperature, the rainfall, and the number of clear days for the five months in ‘each of the last 7 seasons, 1913-1919, are given in Table 1.

The outstanding feature in this table is the high precipita- tion during the growing season. Reference to the data on cli-

WHEAT INVESTIGATIONS. g

matic conditions given by Carlton’* for the principal wheat growing centres in Russia and in the United States shows that the total precipitation for the growing season at Presque Isle— 16.05 inches—is considerably higher than at the principal wheat growing points of the Russian prairies and generally higher than at points in the Great Plains. While this alone would not con- stitute a specially detrimental feature in connection with the cultivation of wheat, a consideration of the peculiarities of the Aroostook growing season indicates that the distribution of the rainfall in the different months of the season may have some effect upon the quality of Aroostook grown wheat. The time of seeding small grains in Aroostook County extends usually

ABI lk

Temperature, Raimfall in Inches and Number of Clear Days for the 5 Growing Months of the Seasons 1913-1919. Re- : corded at Presque Isle, Aroostook County.

Year Temperature | Month Precipitation | | | Nl | No. clear days 1913 | 1914 | 1915 | 1916 | 1917 | 1918 | 1919 | Average | | | | | cee ES ee Pee pes eee eee | | | | | ‘Mean Temp. | 48.60, 53.19) 49.50|50.82 | 48.13) 51.00) 50.50! 49.46 May | Precipitation 3.53) 2.74, 4.05| 3.45 500) 4.00; 3.32! 3.57 INo. clear days | 15 | 18 | 10 {12 biped) Saba oe | Mean Temp. | 56.03] 60.60159.65 | 59.84) | [59108 June Precipitation 1.20} 4.80} 1.95) 2.17 | 7.67 | 1.26] 3.18 Pee clear days 18 WS ea he 10 By S83 3 ‘Mean Temp. 69.30) 63.55 64.10 65.00 67.95) | 65.98 July _| Precipitation | 5.18} 2.23) 3.40] 3.68 2.56 | 3.80 3.48 \No. clear days 9 | 24 1Syee 7, S| 14 16 | | ‘Mean Temp. 61.40, 60.17, 61.80,71.90 | 66.90 61.00, 63.86 August |Precipitation 3.01) 2.385) 3.50} 1.70 5.30} | 1.75] 2.94 Be clear days 20 PBN \) ales pak I< =| | 11 16 ‘Mean Temp. 53.20) 55.80 56.45,58.92 | 53.05) | 58.60) 55.17 September 'Precipitation | 2.01) 2.10) 38.25] 4.05 1.41) | 4.56} 2.90 pies clear days | 19 Ue ft} 19 TS a) 17 | Totals for growing season: | | Total Precipitation | 14.93) 14.22) 16.15|15.05 | 20.84 | 14.69 16.05 Tota! No. clear days 81 95 | 74 |65 70 | | 64 75 Average mean temperature of | | growing season | 58.10| 57.75) 58.49/61.26 | 58.18 55.00 58.70

“Carlton, M. A. The Small Grains, 1916, pp. 699. The MacMillan Co., New York.

10 Marine AGRICULTURAL EXPERIMENT STATION. 1920

from the 8-20 of May. Owing to the cool, moist conditions in Aroostook the vegetative period of wheat is rather extended and the flowering which begins about the middle of July con- tinues to about the end of July. August is the ripening month, the wheat being harvested in the last week of August or the first week of September. Referring to the figures in the last column of Table 1 it will be seen that the highest precipitations occur in May, 3.57 inches, but that the rainfall during the period of ripening and harvest—August,—is only slightly over half an inch or 18 per cent less than the maximum monthly rainfall for the season. Rainy weather during the period of ripening and harvest not only may have a detrimental effect upon the appearance of grain but is associated with another feature of climate, namely humidity which is known to have a distinct effect upon the quality of wheat. High relative humidity as- sociated with an overcast sky characterizes the weather in - Aroostook County towards the end of July and the first week of August which period marks the first stages of the kernel formation. This feature is rather unfortunate as dry weather and a clear sky during the process of ripening are very essential to the production of a strong, high grade wheat This humid condition protracts the ripening period and delays the harvest. The lengthened ripening period extending through August re- sults in a further drawback as the formation of the wheat ker- nel does not coincide under Aroostook conditions with the high- est seasonal temperature, which marks the month of July. While ~ from the seven year average given in Table 1 it appears that the mean temperature in August is only 2 degrees lower than in July, the actual difference is much greater since the amount of sunlight and heat decreases in the shortening days of August. Medium late wheat varieties when planted in the latter part of May often do not mature until the first week of September when frequently the first early frosts occur.

While these are the natural limitations relative to the grow- ing of strong wheat in Aroostook, it must be admitted that the climatic conditions prevailing there are favorable in regard to other features of the wheat crop. Thus the wheat crop is prac- tically free from insect pests, and is seldom affected with stem rust.

WHEAT INVESTIGATIONS. 11 SoILs.

The soils of Aroostook County have been formed by glacial drift, and vary from sand to heavy silt loams. According to Westover and Rowe'’® there are twelve distinct soil types in Aroostook County, but the greater part of the area is made up of a friable loam, Caribou loam, derived from unmodified glacial drift. The Caribou loam is composed of about 50% silt and only 16% of clay, the rest being made up of more or less fine sand and gravel. The soil though most ideal for the potato crop, is well adapted to small grains. The average yield of wheat is about 25 bushels per acre. The fertility of the soil is kept up by high applications of commercial fertilizers in con- nection with the potato crop, very little barnyard manure being used. The humus content of Aroostook soils is restored through cultural methods which consist of a crop rotation usually includ- ing potatoes one year, grain one year and clover and timothy for two or three years. Frequently, however, this rotation is not adhered to potatoes being grown for two or more years in succession. In such cases the drain upon the humus of the soil is probably too great to insure good wheat crops following the potatoes.

CHARACTERISTICS OF THE AROOSTOOK GROWN WHEATS.

The commercial wheat varieties grown in Aroostook are classed with the semi-hard spring wheats of the Northeastern wheat district. The varieties of wheat grown at present in Aroostook belong to two spring wheat groups—Fife and Pres- ton. Owing to the prejudice of some growers against the awned wheat, the majority of the Aroostook wheat varieties belong to the beardless Fife group. More recently the Marquis wheat has found its way from the Northwest into Aroostook County, but does not seem to be so well adapted as the Fife wheats.

An investigation into the physical characteristics and chemi- cal composition of Aroostook grown wheats and into their mill-

**H. L. Westover and R. W. Rowe. Soil Survey of the Caribou Area, Maine, U. S. Dept. Agr. Bur. of Soils, 1910, pp. 1-40.

a TN By met

12 MAINE AGRICULTURAL EXPERIMENT STATION. 1920

ing and baking value was made by Woods and Merrill”. An especially interesting feature of their investigation in its bearing upon the observations reported in the present paper is their study on the effect of the climate upon the physical appearance and chemical composition of wheats imported from the North- west and grown in Aroostook. They found that the wheat vari- ties introduced from the Northwest changed their physical and chemical character at the end of a single season, the change being most pronounced in the increased size of the kernel. Rela- tive to the effect of Aroostook environmental conditions upon | the protein content of Northwestern wheats the evidence from the trials of Woods and Merrill is inconclusive. In the first experiment one of the three varieties tested, Lamona wheat, suffered at the end of the first season under Aroostook condi- tions a large loss of nitrogen and a still larger loss in the gluten content, while the Fife wheat showed only a small decrease in gluten and the Bluestem wheat made a slight gain in protein, both these varieties gaining in gluten. The results from the second experiment in which two pure strains of Minnesota bred Bluestem and one commercial variety of Bluestem were used, _ showed a decrease in protein content for the commercial variety and for one of the pure strains as compared with the Minnesota grown parents, but on the following year all three varieties showed a higher percentage of protein in the Aroostook grown progeny than in the check trials of the Minnesota grown pro- geny.

As to the baking quality of the Aroostook grown wheats the baking tests reported by Woods and Merrill show that the flour from Maine wheats produced as a rule loaves of smaller volume than the Minnesota standard flour, though of good quality. The writers noting the arbitrary nature of the north- western standard, suggest that Maine develop the growing and milling of wheat along its own lines, and express the belief that “by careful breeding from wheat now being grown in Maine it would be possible to develop a strain equal for Maine conditions to some of the improved strains of other sections.”

“Chas. D. Woods and L. H. Merrill. Notes and Experiments upon the Wheats and Flours of Aroostook County. Maine Agr. Expt. Sta. Ann. Rept. 1903, pp. 145-180. (Bull. No. 97).

WHEAT INVESTIGATIONS. 13 MATERIAL AND METHODS.

In undertaking the wheat improvement work at Aroostook Farm in 1915 the question arose as to what material should be used as the source of improved wheat strains. A consideration of the deterioration in the physical characteristics of the north- western wheats under Aroostook conditions, and of the great differences between the environment of Maine and the North- west, at once suggested the advisability of confining the selection work chiefly to the native Aroostook grown wheats or adjoining regions. Since the quality of the wheat crop appears quite susceptible to the influence of climatic factors it was thought that the reaction, if any, of the different varieties and strains in the course of many seasons to the environmental factors has long become established in the form of a greater or less degree of adaptation. Selection work on these varieties would result in the isolation of the best adapted varieties or strains and in the elimination of the poorly adapted ones, the degree of adap- tation being measured by the maximum quality of any given strain.

The methods used in the wheat improvement work at Aroostook are based on the principles of pure line selection and hybridization. The present paper deals only with the results of the selection work. For a detailed account of the method of pure line selection and of the field technic as applied to small grains by this Station the reader may be referred to a previous paper.** We may only consider here a few features not men- tioned in the paper just cited, which are peculiar to wheat. In selecting wheat strains for quality a certain procedure of diag- nostic value is required by means of which the relative quality of the different individual plants may be determined. The yield of grain from a single spike is obviously too small to be used for a nitrogen analysis or even for a gluten determination by the chewing test without interfering with the propagation of the seed. The estimation of the quality of the grain from the individual spikes following their isolation from commercial vari- eties was based in this work upon the hardness, color, size and

“Frank M. Surface and Jacob Zinn. Studies on Oat Breeding IV. ‘Pure Line Varieties. Maine Agric. Exp. Station, Ann. Rept. 1916, pp. 97-148 (Bulletin No. 250).

14 Maine AGRICULTURAL EXPERIMENT STATION. 1920

texture of the kernel. While it is recognized that these fea- tures are not always a reliable index of the quality of wheat, especially when estimating grains from different varieties, yet the diagnostic value of these determinations is enhanced when they are made upon different strains within the same variety. The determination of the physical characters of a number of strains within the same variety soon leads to the formation of a standard for each variety so that the relative quality of the ker- nels from the strains of the same variety can be fairly accurate- ly judged. It may be added in this connection that the deter- mination of the characters of the grain at the early stage of selection is really of no great importance since it is the progeny of the selected plants, the first generation after selection, that offers a more reliable basis for the determination of quality.

BRIEF ACCOUNT OF THE ORIGIN OF THE PURE LINES OF WHEAT AT ARoosTOOK FARM.

In 1915 several hundred selections were made from com- mercial varieties of wheat grown at Aroostook Farm and from a number of wheat fields in the County. In this work normal, medium sized spikes well developed at the tip, were selected rather than whole plants since in the close field stand it is not always possible with certainty to isolate individual plants. In these selections were represented wheat spikes of the four groups of spring wheat: Fife, Bluestem, Preston and Durum. Representative spikes and grain of these wheats are shown in Figures I to 6. Of these selections 259 spikes were retained and planted in rows in the cereal crop nursery in 1916. Each row was planted with the seed of a single wheat spike. The number of rows grown in 1916, each representing a strain selected from the different varieties is given on page 17.

During the growing season notes were taken on the char- acters of the spikes, tillering capacity, strength of straw, sus- ceptibility to disease as well as data relative to time of heading and bloom. The data on the physical characters of the grain in conjunction with the field notes served as a basis for further selection as a result of which g1 strains of the original 259 were retained. These 91 strains were planted in 1917 in one two- thousandth acre plots. Along with the pure lines of Aroostook

WHEAT INVESTIGATIONS. 15

Figure 1. Ficure 2.

Ficure 3. é Figure 4.

Fic. 1.—Representing heads of two groups of spring wheat: A, Red Fife (Line No. 2393); B, Bluestem (Line No. 2387).

Fic. 2.—Heads representing two spring wheat varieties: A, Marquis (Line No. 2398); B, Preston (Line No. 2388).

Fic. 3.—A, Royalton Red, Accession No. 186; B, Royalton White,

Accession No. 185. Fic. 4—Head of Durum wheat (Speltz Marz, Accession No. 182.)

16 MAINE AGRICULTURAL EXPERIMENT STATION. 1920

Variety No. of Garden Rows _ No. of Strains Con-

in 1916 tinued in 1917 Red Fife 41 15 Preston 37 17 Bluestem 25 20 Marquis 49 12 Canada Red* 32 19 Unnamed 70 8 Durum 5 = Total 259 91

grown wheats there were also tested in one two-thousandth acre plots 7 pure strains of hard spring wheat which had been ob- tained from Minnesota in the winter of 1916. The writer wishes to express his appreciation of the courtesy of Prof. H. K. Hayes of the University of Minnesota in sending these wheats to him. The Minnesota strains included one representa- tive of each of the following varieties: Bluestem (Haynes Blue- stem), Marquis, Velvet Chaff, Royalton (Red), Royalton (White) and two representatives of durum wheat, Speltz Marz (Fig. 4) and Hedge Row. According to a written communica- tion from Professor Hayes the Royalton wheat was originally obtained from Royalton, Minn., and its origin was possibly a natural cross. The two strains of this wheat are very distinct, one (Royalton White) possessing a smooth chaffed spike and white grain suggestive of the White Fife, while the other strain (Royalton Red) produces a red grain and a hairy chaffed spike similar to Bluestem. The spikes of these two strains are repro- duced in Fig. 3. All the Minnesota strains have been given Maine accession numbers 182 to 187.

The crop from the one two-thousandth acre plots in 1917 furnished enough seed from each strain to make a chemical analysis of the crude protein content. A further scrutiny of these pure lines on the basis of the chemical analysis and the field notes resulted in the discarding of a number of strains until 44 pure lines of Aroostook wheats and 6 pure lines of

*Dr. Chas. E. Saunders, Cerealist at the Central Experimental Farm: at Ottawa, informs me that the variety here listed under the name Can- ada Red is probably the variety called Black Sea (identical with Ladoga). It is a bearded wheat, with smooth, reddish brown chaff and producing red kernels.

WHeEAT INVESTIGATIONS. W,

Minnesota wheats were retained. These were grown in the sea- son of 1918 in plots ranging from one two-hundredth to one- fortieth acre in area. A further selection in the fall of 1918 reduced the number of strains to 26 originally selected from Aroostook wheats and to 5 Minnesota lines. A complete chem- ical analysis of 37 lines and baking tests of 31 wheat lines were made in the spring of 1919. Using the baking test as the final index of strength, out of the 31 tested strains 12 Aroostook lines

and 4 Minnesota lines were retained and propagated in 1919 in.

one-twentieth acre plots.

ANALYSIS OF DATA.

PHYSICAL CHARACTERISTICS OF THE GRAIN.

The most striking feature in the physical appearance of the grain from the Aroostook grown lines is the size and weight of the kernels. Their size is larger than that of spring wheat of the Northwest and their plump, well rounded shape, distin- guishes them from the wheats of the prairie regions. The size and shape relations of the grain of the Aroostook pure lines are illustrated in Figures 5 and 6, which represent kernels of four pure lines each belonging to a distinct wheat variety. As a re- sult of these size and shape relations the weight of the kernels is correspondingly high. In Table 2 is given the weight per 1000 kernels of each of the pure lines grown in 1918. The lines are grouped according to the variety from which they originated and their rank in respect to kernel weight.

It may be well to point out that the determination of the kernel weight based on 1000 kernels expresses very exact rela- tions repeated determinations showing the experimental error to be either negligible or nil. Therefore, the differences in the 1000-kernel-weight of these lines are more significant than they might appear at first sight. From Table 2 it will be noted that the average weight per 1000 kernels for all lines is quite high—35.314 grams. ‘The average weights per 1000 ker- nels for the varieties do not show great differences though they very well reflect varietal means around which the strains with- in the varieties are grouped. The individual strains show a considerable deviation from the average of the varieties from

se eee _—

18 - Marne AGRICULTURAL EXPERIMENT STATION. 1920

Fic. 5—Kernels of two pure lines of wheat: A, Line No. 2389 (Red Fife) ; B, Line No. 2384 (Bluestem).

19

WHEAT INVESTIGATIONS.

v

nn

o

t=

Ay 4 (oe)

©O

Se)

a 5 fo)

Za

o S ; |

xq

B

o pe}

5 Ore

° ae Sl Ho 23 pe OW a 33 on dele G ore ga Sle LS ‘o n~ og A IH [oe OG

ton)

20 MaINneE AGRICULTURAL EXPERIMENT Station. 1920 ANE Weight per 1000 Kernels of each Pure Line.

AROOSTOOK LINES

Line No. Selected from Variety Weight in grams 2389 Red Fife 31.180 2385 Red Fife 34.093 2379 Red Fife 34.344 2386 Red Fife } 34.358 2395 Red Fife 34.535 2390 Red Fife 34.637 2393 Red Fife 34.882

Average 34.004 2394 Bluestem 34.160 2391 Bluestem 35.102 2384 Bluestem 36.043 2387 Bluestem 37.460

Average 35.691 2383 Preston 26.541 2397 Preston 30.186 2388 Preston 32.535 2414 Preston 33.556 2492 Preston 39.093 2405 Preston 43.899 2409 Preston F 44.032 2400 Preston 44.789

Average 36.828 2410 Marquis 33.174 2398 Marquis 34.055

Average 33.611 2373 Canada Red 30.185 2396 Canada Red 34.498

Average 32.337 2370 Unnamed 35.942 2415 Unnamed 37.341 2406 Unnamed 37.574

Average 36.952

Average for all Aroostook lines: 35.314

which they originated as well as from the average for all lines. Thus the Preston Line 2383 has a 1000-kernel-weight over Io grams lower than the average for the variety and over 18 grams lower than the highest weight in this variety, —44.789 grams of Line 2400. This line exceeds the average for the variety by nearly 8 grams. With the exception of Line No. 2389 the Red Fife lines show a remarkable uniformity in the 1000 kernel weight with an average below that for all lines. The Bluestem strains rank higher with an average equal to that for all lines,

WHEAT INVESTIGATIONS. 21

while the Preston lines, with the exception of the few strains of the Unnamed variety, rank highest. The great variations in the tooo-kernel-weight of the individual strains grown in the same season on a limited, uniform piece of land, are rather note- worthy. It is further of interest to note that the lines with the very high weight per 1000 kernels as a rule yielded flour of poor baking quality. The lines that produced the largest bread volume were those with a 1000-kernel-weight not over 35 grams.

Table 3 shows the remarkable effect exerted by the Aroos- took environment upon the weight of kernels of the Minnesota lines.

iba ey Bs)

Showing Effect of Aroostock Growing Conditions Upon the Weight of Kernel of Pure Lines Introduced from Minnesota.

. Weight in grams per 1000 kernels Maine Minnesota ; ed Acees- Accession |N.S.N. of original seed | of the same seed sion No. No. Selected from grown in Minne- | grown one season Variety ; sota in Aroostook | j 183 | 1011 T-15-161 IVelvet Chaff ; 19.785 32.891 186 1037 I-12-1 Royalton (Red) | 22.000 | 34.315 185 1037 | I-12-6 Royalton (White) | 20.923 35.105 181 188x188 1-06-39 Preston x Preston | 22.268 37.241 184 470 1-06-52 [Hedge Row (durum) 37.411 43.104 182 337 1-00-45 Speltz Marz (durum) 37.386 | 45.510 Average 26.629 38.028 Average (excluding the durum lines) 21.244 34.888

The average weight per 1000 kernels of the original seed has increased by 18.7 per cent in the course of one season. It should, however, be stated that at the time of the determination of the rooo-kernel-weight the original Minnesota grown seed was older and therefore drier than the Aroostook grown seed and considerably shriveled, so that the actual difference in weight should probably be smaller. On comparing the data in Table 3 with those in Table 2 it will be seen that the average weight per 1000 kernels of the Aroostook lines is practically identical with the average weight of the Minnesota lines attained under Aroostook conditions. These data which very well agree with

are

22 Matne AGRICULTURAL EXPERIMENT STATION. 1920

the observations of Woods and Merrill’® on 2 races of Minnesota wheat are too meager to draw from them general conclusions.

The evidence obtained from the examined number of pure lines.

indicates that the environmental factors prevailing in Aroostook

brought the kernel weight of the original Minnesota strains up to the level of the Aroostook strains very rapidly, in fact, within.

a single season. Further determinations made on the grain of

the crop of 1918 indicate that this change in the kernel weight

of the Minnesota lines is not progressive.

In regard to other physical characteristics as color, texture and hardness it was found that there was some variation among

the different lines, but only of a comparatively slight nature since these characteristics were primarily used as a basis for selection. As a rule the strains with the highest weight per 1000 kernels

showed the least degree of flintiness and the highest percentage

of “yellow berry”. Of the Minnesota wheats 4 strains Royalton (Red), Royal-

ton (White) Haynes Bluestem and Marquis, respectively, while

suffering some loss in flintiness as compared with their original condition in Minnesota, appeared to be of good color, and tex- ture; of the remaining three Minnesota strains the Velvet Chaff strain showed a tendency towards developing “yellow berries” while the two durum lines showed a most striking degree of de- terioration. The original Minnesota sample of these lines ex- hibited all the characteristics of the corneous transclucent grains of the Northwestern durum wheats; after one season’s growth under Aroostook conditions the already large grain gained from

6 to 8 grams per 1000 kernels and showed a very large percent-

age of opaque kernels of either partly or wholly starchy texture. This rapid change of the durum strains grown side by side with

a number of other lines retaining their hard texture and good color furnished the best illustration of the difference in the de- gree of adaptation and response to the environment of the dif-

ferent wheat varieties. The effect of the Aroostook environment

upon the size and shape of the original Minnesota seed is illus-

trated in Figures 7 and 8.

ALCEs Cte

WHEAT INVESTIGATIONS. 23

Fic. 7—Showing effect of Aroostook environmental conditions upon the size and shape of wheat introduced from Minnesota: A and B, origi- nal, Minnesota grown seed (Royalton Red and Royalton White, respec- tively); A’ and B’ same seed after one esason’s growth at Aroostook

Farm.

24 MaINeE AGRICULTURAL EXPERIMENT STATION. 1920

Fic. 8.—Showing effect of Aroostook environmental conditions upon the size and shape of wheat introduced from Minnesota: A and B, origi- nal, Minnesota grown seed (Marquis and Durum, respectively) ; A’ and B’ same seed after one season’s growth at Aroostook Farm.

WHEAT INVESTIGATIONS. 25

YIELD OF THE PurE LINES IN 1918.

As already stated the pure lines were propagated in 1918 in plots ranging according to the available seed from one two- hundredth to one-fortieth acre in area. The season of 1918 was typical of Aroostook conditions, marked by heavy rain-storms which lodged some of the wheat thus offering an opportunity of

TABLE 4.

Yield and Weight per Measured Bushel of Pure Lines Grown im IOI. AROOSTOOK LINES.

Yield of Grain

| Line} Selected from Weight per No. | Variety | Plot No. Pounds Bushe!s measured | per plot | per acre bushel in Ibs. | | | | | | 2385 | Red Fife 870 5.13 | 51.58 60.5 2395 Red Fife 882 12.69 | 39.88 | 61.5 2393 | Red Fife 880 11.44 38.62 61.3 2386 | Red Fife 871 12.25 37.73 62.0 2379 | Red Fife 864 4.50 33.72 2389 | Red Fife 875 10.00 | 33.51 61.0 2390 | Red Fife 876 8.06 26.15 60.5 | | 2387 | Bluestem | 873 | 11.88 48.00 | 60.0 . 2404 | Bluestem | 890 10.69 34.66 2384 | Bluestem . | 869 10.81 38.92 60.5 2394 | Bluestem 881 11.00 | 34.54 60.0 2391 Bluestem | 877 10.31 33.44 | 60.5 2382 | Bluestem 867 3.00 23.93 | 2402 | Preston | 893 3.50 53.79 2414 | Preston 910 39.75 40.04 62.5 2405 | Preston 891 | 13.34 | 39.26 60.5 2400 | Preston | 907 40.00 | 34.62 | 62.0 2409 | Preston } 900 | 20.88 32.48 62.0 9397 | Preston 884 7.75 | 31.79 | 63.3 2388 | Preston | 874 9.00 31.10 | 63.0 2381 | Preston | 866 | 4.34 | 29.25 2380 | Preston | 865 | 1.65 28.62 | 2383 | Preston | 868 5.25 25.20 | 61.5 2411 | Preston 902 14.00 | 25.41 | 59.8 2408 | Preston 894 7.13 7 BBY Al 2398 | Marquis | 885 10.38 35.27 | 61.3 _ 2410 | Marquis 895 11.50 | 25.73 62.0 2378 | Canada Red | 909 5 17.00 42.74 2375 | Canada Red 904 24.00 39.35 2376 | Canada Red 905 20.00 38.54 2374 | Canada Red | 991 22.75 | 36.47 2372 | Canada Red | 898 22.00 36.68 2371 | Canada Red 896 21.75 35.73 2373 | Canada Red | 899 21.50 35.04 57.5 2396 | Canada Red | 883 8.81 33.80 62.0 | 2406 | Unnamed S27 til 13.94 44,49 57.5 2415 | Unnamed | 911 31.00 | 43.55 60.0 2370 | Unnamed | 862 4.75 | 37.54 | 60.5

26 MAINE AGRICULTURAL EXPERIMENT Station. 1920

Yield and Weight per Measured Bushel of Pure Lines Grozwn im 1918.—Concluded.

MINNESOTA LINES.

| -Yield of Grain

137 1239 Marquis 88s | 6.56 35.81 60.5 182 337 Speltz Marz

(durum) 878 11.25 | 34.62 64.5 18 1911 Velvet Chaff | sso | 10.25 34.29 69.5 186 1037 Royalton (Red) S72 10.09 34.17 60.8 189 169 Haynes Bluestem | 879 11.00 | 33.89 59.8 185 1037 Roya'ton (White) 897 13.13 | 32.59 60.5

judging the relative strength of straw of the different lines. The smaller plots were cut by hand with the sickle or cradle, the lar- ger ones were reaped with the binder. The threshing was done according to the size of the plots, either with the small threshing machine used in threshing experimental plots up to one-eightieth acre in area, or with the large, farm threshing machine. The yields of 44 pure lines and the bushel weight of most of them are given in Table 4.

While relatively considered, these plots are of different size and absolutely taken, quite small, there were enough plots of ap- proximatly the same area to give some idea of the behavior of. these lines as to yield. An inspection of Table 4 will show a con- siderable variation in the yield of the different lines. Every vari- ety contains strains of high and low yielding capacity. In the cases where a larger yield was obtained from an equal or smaller area the differences in yield are probably significant. Consid- ering the larger plots in which the Canada Red strains grew, this variety furnished a number of high yielding lines. Some of the Preston lines yielded well, as did some of the Red Fife and Bluestem lines. The Minnesota lines very well approached the yielding capacity of the Aroostook lines.

From Table 4 it will be noted that these lines test rather high and that there is little variation in the bushel weight. Only two of the 44 lines tested appreciably below the standard weight, the great majority exceeding it by 0.5 to 3.pounds, and in one case (durum) by 4.5 pounds. The Preston lines stand out rather prominently with their higher bushel weight, and it is of interest

WHEAT INVESTIGATIONS. TH,

to note that line 2388 showing the highest bushel weight of all Aroostook lines furnished a flour of very good quality, best of all Preston lines. The Minnesota lines, with the exception of durum, are well grouped about the standard weight.

CHEMICAL CHARACTERS OF THE PuRE. LINEs. PROTEIN ANALYSIS IN IQI7.

The protein content of the pure lines was first determined in 1917.* In computing the protein content from the nitrogen the factor 5.7 was used. The results of the analysis of g9 pure lines are given in Table 5. The lines are grouped within the varieties from which they originated, and according to their nitrogen rank.

An inspection of Table 5 shows a relatively high average percentage of protein for the different varieties. Certain lines within each variety group possess a protein content which under Aroostook conditions must be regarded as decidedly high. This should not surprise one if it is recalled that the lines were selec- ted for high nitrogen as indicated by their physical characteris- tics. From Table 5 it will be noted that the Canada Red variety shows the highest average protein content —14.31 per cent—of all Aroostook lines. This group 1s followed by the Preston, Blue- stem, Red Fife, Unnamed and Marquis varieties in the order named. The Minnesota lines, excluding the durum wheats which proved to be least adapted to Aroostok conditions and should be treated separately by themselves, show an average crude protein content practically equal to that of Canada Red. The average protein content of all Aroostook lines taken together is 13.81 per cent while the Minnesota lines with an average of 14.36 show a hardly significant difference of about 0.5 per cent.

It will be of interest to compare the protein content of the Minnesota lines originally grown in Minnesota with that .of their progeny grown one season in Aroostook. In Table 6 are tabulated the data on the protein content of the original Minne- sota grown seed and of the Aroostook grown progeny.

*Credit is due to the Chemistry Department of this Station for the _ chemical analyses of wheat of the pure lines in 1917 and 1918.

pee

28 MaIne AGRICULTURAL EXPERIMENT STATION. 1920

IUNBILID,

Crude Protein Content of Pure Lines of Wheat Grown at Aroostook Farm im 1917.

AROOSTOOK LINES

WuHeEat INVESTIGATIONS. 29

Crude Protein Content of Pure Lines of Wheat Grown at Aroostook Farm in 1917.—Concluded.

Line No. Selected from | Crude Protein Variety Plot No. Nitrogen (N x 5.7) Per cent |

Average ; | 2.45 13.94 2410 Marquis 620 2.58 14.71 2556 Marquis | 622 2.45 138.97 2555 Marquis | 621 2.41 13.74 2587 Marquis 630 2.36 13.45 2583 Marquis 629 2.34 13.34 2401 Marquis 632 2.30 13.11 2398 Marquis 624 2.29 13.05 2558 Marquis 623, 2.29 18.05 2561 Marquis 625 2.26 12.88 2588 Marquis 631 2.26 12.88

- 2579 Marquis Q 628 2.24 12.77 2569 Marquis 627 2.20 12.54

Average 2.33 13.29 2374 Canada Red 612 2.76 15.73 2377 Canada Red 613 2.64 15.05 2378 Canada Red 609 2.62 14.93 2373 Canada Red 604 2.61 14.88 2376 Canada Red 608 2.61 14.88 2372 Canada Red 605 2.61 14.88 2375 Canada Red 614 2.58 14.71 2544 Canada Red | 618 2.56 14.59 2371 Canada Red 619 2.54 14.48 2543 Canada Red 617 2.50 | 14.25 2535 Canada Red 610 2.50 | 14,25 2531 Canada Red 607 2.46 14.02 2529 Canada Red 606 2.44 13.91 2523 Canada Red 601 2.42 | 13.79 2524 Canada Red 602 2.40 | 13.68 2522, Canada Red 600 2.38 | 13.57 2525 Canada Red | 603 2.26 12.88 2536 Canada Red | 611 2.24 12.77

cant TE IPO i Pc eae er Average | | 2.40 13.69

MINNESOTA LINES.

183 1011 [-15-161 Velvet Chaff | 654 2.71 15.45 189 169 1-16-19 Haynes Bluestem | 648 252, 14.36 186 1037 1-12-1 Royalton (Red) | 651 2.48 14.14 185 | 1037 1-12-6 Royalton (White) | 647 2.48 =| 14.14 181 | 188x188 TI-06-39 |Preston x Preston | 652 2.47 | 14.08. 187 1239 1-16-81 Marquis 650 2.46: } 14.02 182 337 1-00-45 Speltz Marz (durum)) 653 2.18 12.43 184 470 I-00-52 Hedge Row (durum) 649 2.04 11.59 Average 2.42 13.79 Average for Minnesota Lines (exclusive of durum lines) 2.52 14.36 Average for Aroostook Lines 2.43 13.81

30 MAINE AGRICULTURAL EXPERIMENT STATION. 1920

A comparison of the two sets of data given in Table 6 shows that the Minnesota lines when transferred into Aroostook and grown there one season suffered a loss of only about 0.8 per cent of protein. If we further add that the original Minnesota grown seed, when’ analyzed for protein was considerably older, and having a lower moisture content than the Aroostook progeny grown from it, the latter’s loss of 0.8 per cent of protein as com- pared with the former becomes negligible.

TABLE 6.

Comparison of the Proteim Content of the Minnesota Grown Lines with that of their Aroostaok Grown Progeny.

Original! Minnesota’ Aroostook grown

We cent | per cent | per cent per cent

| | 182 337 45 Speltz Marz (auim)| | 15.68 | 138) | 12.43

T-00- 2.75 2.1. 186 1037 I-12-1 Royalton (Red) 2.72 15.50 2.48 14.14 183 1011 I-15-161 Velvet Chaff 2.71 15.45 2.71 15.45 184 470 1-00-52 Hedge Row (durum) 2:68. 1) 1528 2.04 11.59 185 1037 I-12-6 ‘Royalton (White) 2.55 -| 14.54 | 2:48 14.14 | Average 2.68 | 15:30 | 2.42, 13.79 Average (excluding durum lines) 2.66 15.17 V5) 14.36

A consideration of the changes caused by the Aroostook environment in the weight, size, shape of kernel as well as in the protein content of the original Minnesota grown seed very clearly indicates that the effect of the Aroostook environment upon the Minnesota grown wheats is decidedly more noticeable in the physical characteristics than in the crude protein content. A very notable exception are the two durum lines. The Minne- sota grown Speltz Marz heads the list of the Minnesota lines in regard to protein content; at the end of one season’s growth in Aroostook this line shows the second lowest protein content of all 99 lines analyzed. The second durum line, Hedge Row, suf- fered an even greater loss in nitrogen —3.69 per cent. By con- trast, the Aroostook grown Velvet Chaff line shows a protein content identical with that of the Minnesota grown seed, while

WHEAT INVESTIGATIONS. 31

Royalton (White) and Royalton (Red) suffered only a loss of 0.40 and 1.36 per cent, respectively. The grain of the durum lines showed also the most marked deterioration in the physical characteristics. This tendency of the durum wheats to rapidly deteriorate under Aroostook conditions was also borne out by the results from the 1918 crop which will be considered below.

A comparison of the average protein content of the Aroos- took wheat lines with the average protein content of the chief classes of American wheats as given by L. M. Thomas”® is of interest. In his extensive investigations made in the Office of Grain Standardization and involving several hundreds of samples, Thomas found the average crude protein content of the soft red winter, durum, hard red winter, and hard red spring wheats POO Onno) 121. L2.ONper cent, respectively. dhe average crude protein content of the Aroostook pure lines of wheat grown in 1917 was found to be 13.81 per cent or nearly 1 per cent high- er than the average for the class of hard red spring wheats as given by Thomas.

CHEMICAL ANALYSIS OF THE PurE LINES GROWN IN 1918.

At the stage of propagation of the pure lines reached with the harvest of the 1918 crop there was enough seed to make a complete chemical analysis and baking tests of these wheats. Of the 99 lines grown in 1917 only 44 were continued in 1918 and of these 40 were subjected to a chemical analysis. This reduction was due to the elimination of strains lacking in pro- tein or in strength of straw and other desirable physical characteristics. The results of the chemical analysis of the 40 lines are presented in Table 7. The lines within each variety are grouped according to their protein content. _ A study of the data in Table 7 will reveal first a general lowering of the average protein content of the pure lines as com- pared with the protein content of the 1917 crop. The variation in the average protein content ranges from 11.76 to 13.30 per cent as against 13.29 to 14.36 per cent for the 1917 crop when

“Thomas, L. M. A Comparison of Several Classes of American Wheats and a Consideration of Some Factors Influencing Quality. U.S. Dept. Agr. Bu. of Plant Ind. and Bu. of Markets 1917, Bul. 557, pp. 1-28.

32 Matne AGRICULTURAL EXPERIMENT STATION. 1920

the range of variation was somewhat narrower than in 1918 The growing conditions of the 1918 season had admittedly a levelling down effect upon the protein content of the pure lines. Apart from the seasonal influences there were other factors in

TABLE 7.

Chemical Composition of Pure Lines of Wheat Grown at Aroos- took Farm im 1918.

AROOSTOOK LINES.

] Line | Selected from Plot |Moisture Ash Nitrogen Protein | Fat Crude No. | Variety No. » (Nx5.7) Fiber 2385 Red Fife 870 7.13 1.92 2.55 14.54 2.64 2.18 2390 Red Fife 876 7.98 2.05 2.42 13.79 2.42 2.49 2389 Red Fife 875 8.63 2.04 2.25 12.83 2.54 2.04 2393 Red Fife 880 ; 9.88 1.98 2.14 12,20 2.01 2.10 2379 Red Fife 864 8.51 2.03 2.14 12.20 2.57 2.19 2386 Red Fife 871 9.65 1.95 2.13 12,14 2,24 2.19 2395 Red Fife 882 9.17 1.92 2.06 11.14 | 2.61 2.09

Average 8.71 1.98 2.24 12.77 | 247 | 2.18 2404 Bluestem 890 7.28 2.07 2.36 13:45 wel ion 2.25 2391 Bluestem | 877 7.52 2.01 2.34 13.34 2.38 2.26 2394 Bluestem | 881 7.17 2.07 - 2.22 12.65 2.57 2.27 2387 Bluestem | 873 8.74 2.16 2.19 12.48 2.56 2.20 2384 Bluestem 869 9.95 2.01 2.09 11.91 2.30 2.18

Average 8.08 2.06 2.24 12.77 2.50 2.23 2381 Preston 866 7.88 2.15 2.54 14,48 2.95 2.19 2402 Preston 893 8.71 2.12 2.42 13.79 2.54 2.36 2411 Preston 902 7.33 2.24 2.42 13.79 2.59 2.23 2397 Preston 884 8.58 2.11 2.34 13.34 2.53 2.16 2383 Preston 868 8.03 2.29 2.32 12:22 2.21 2.44 2414 Prseton 910 8.70 2.06 2.29 13.94 2.31 DAG 2388 Preston 874 9.17 1.96 2.26 12.88 2.36 2.23 24909 Preston 900 9.83 1.99 2.09 11.91 2.42 2.09 2405 Preston 891 9.21 2.08 2.02 11.51 2.88 2:33 2400 Preston 907 9.24 2.01 1.93 11.00 2.38 2.19

Average 8.67 2.10 2.26 12.91 2.52 2.24 2398 Marquis 885 7.83 2.16 2.12 12.08 2.46 2.53 2401 Marquis 887 9.05 1.87 2.09 11.91 2.38 1.99 2410 Marquis 895 7.77 1.99 1.98 11.29 2.75 2.28

Average 8.22 2.01 2.06 11.76 2.53 2.27 2374 Canada Red 901 8.05 2.35 2.40 13.68 2.70 2.41 2377 Canada Red 908 8.44 2.16 2.37 13.51 2.20 2.54 2373 Canada Red 899 ” 8.46 2.14 2.35 13.40 2.63 2.53 2376 Canada Red 905 7.61 2.04 2.33 13.28 2.28 2.68 2375 Canada Red 904 7.93 2.09 2.12 12.08 2.50 2.49 2396 Canada Red 3 9.85 2.12 2.08 11.86 2.50 2.34

Average 8.98 2.16 2.32 13.20 2.46 2.53 2370 Unnamed 862 8.28 1.95 2.28 12.99 2.25 2,24 2415 Unnam:d 911 8.29 1.91 2.16 12.31 2.49 2.23 2406 Unnamed 892 8.89 1.90 2.03 11.57 2,32 2.39

Average 8.49 1.92 2.16 12.29 2.35 2.29

= ee

WueEat INVESTIGATIONS. 33

Chemical Composition of Pure Lines of Wheat Grown at Aroos-

took Farm in 1918.—Concluded. MINNESOTA LINES.

| Accession Selected from | Protein Fat | Crude No. Variety | Moisture} Ash | Nitrogen) (Nx5.7) Fiber 187 Marquis 8.04 2.01 2.44 13.91 2,85 | 2.23 186 Royalton (Red) | 40% | 2.08 2.38 13.57 DATES — )| PABAl 185 Royalton (White) 9.44 | 2.02 2.32 13.22 2.31 2.23 180 Haynes Bluestem oe efes3 2.05 2.27 12.94 2.74 2.22 183 Velvet Chaff | 9.20 1.97 2.26 12.88 2.46 2.14 182 Speltz Marz (durum) 8.80 1.93 1.83 10.43 2.50 2.08 Average for Minnesota lines 8.28 2.03 225) 12.83 2.60 2.19 Average for Minnesota lines - except durum 8.18 2.03 2.33 12.30 2.62 2.21

regard to which the crops of 1917 and 1918 differed. In 1917 the wheat lines did not grow on typical Caribou loam but on a darker soil with more abundant moisture and possibly more hu- mus, which may have accounted for the higher protein content in 1917. Further, in 1917 the wheat lines all grew in one two- thousandth acre plots while in 1918 the area ranged from one two-hundredth to one-fortieth acre. The smaller tract occupied by the wheats in 1917 presented probably a greater uniformity of soil conditions than the larger area in 1918 which possibly accounted for a narrower range of variation in the protein con- tent.

It will now be of interest to study the relationship between the protein content in 1917 and 1918. Considering first the va- rieties as a whole we note on consulting Tables 5 and 7 that not- withstanding the comparatively small differences in their pro- tein content the varieties rank practically in the same order with respect to protein content in 1918 as they did in 1917. This is brought out more clearly by bringing together the average pro- tein contents of the pure lines of each variety for each year as shown in Table 8. While the difference between the averages are small the data given in Table 8 indicate a tendency for the varieties as a whole to preserve their respective rank with re- spect to protein content.

As the average of these varieties are determined by the val- ues of their component strains it will be instructive to examine the behavior of the individual lines with respect to their protein content from year to year. It will be remembered that 99 lines.

ST

34 MAINE AGRICULTURAL EXPERIMENT StTATIon. 1920 TABLE 8.

Relation Between the Average Protein Content for the Pure Lines of Each Variety in 1917 and 10918.

A Average Protein Content Parent Variety in per cent of Pure Lines :

1917 1918 |

Minnesota Lines 14.36 13.30 Canada Red 14.30 13.20 Preston 13.94 12.91 Bluestem | 13.74 12.77 Unnamed | 13.69 12.29 Red Fife 13.68 12.77 Marquis 13.29 11.76

were analyzed in 1917. In 1918 only 44 lines were continued and of these 40 were analyzed so that only for these 4o lines could the variation in the protein content in 1917 and 1918 be deter- mined. The protein content of each line in 1917 was correlated with the protein content of the same line in 1918 with a resulting correlation coefficient of 0.381-+.092. This coefficient as judged by its probable error is significant despite the small number of in- dividuals, and indicates that with a number of pure lines here considered the protein content was transmitted from one year to the next. This is in accord with the observations reported by Freeman*? who found for the character hardness or flintiness of wheat which is regarded as closely associated with high nitro- gen, that the “differences (in percentage of hard grains) were varietal and tended to persist in the same strains from year to year.” Freeman’s pure lines were selected from a commercial variety of Turkey wheat and the percentage of hard grains in 1914 correlated with percentage of hard grains in 1915==57% +4% ; the percentage of hard grains in 1915 with percentage of hard grains im) 1O16—23,9o=-5 7o( Picep a 2y)e

The degree of correlation between the protein contents of wheat varieties from one year to the next will depend upon the number of strains which under a given set of environmental con- dititions will tend to’ retain their relative rank with respect to

“Freeman, Geo. F. A Mechanical Explanation of Progressive Changes in the Proportions of Hard and Soft Kernels in Wheat. Jour. Am. Soc. Agr. 1918, v. 10, pp. 23-28.

Wueat INVESTIGATIONS. 35

protein content. Since the coefficient of correlation referred to above was obtained by grouping pure strains originally selected from different varieties it became a matter of interest to deter- mine to what extent each of the respective parent varieties in- fluenced the value of that coefficient. A tabulation of the differ- ent strains within each variety with respect to their relative rank in protein content in 1917 and 1918 brought out the fact that the varieties Canada Red, Preston, Red Fife and Unnamed furnished most of the strains which retained their relative rank in protein content from one year to the next, while with the Minnesota lines and the Marquis and Bluestem varieties the behavior of the lines ‘with respect to this character was more erratic. It may be of interest to note in this connection that the varieties Preston, Can- ada Red and Red Fife which contained a large number of lines retaining their relative rank in protein content from year to year have been grown for years in Aroostook and Eastern Canada whereas the other varieties whose lines did not show a consis- tent behavior with respect to protein content have only very re- cently been introduced into Aroostook. This fact may serve to explain the different behavior of these two groups of strains with respect to their protein rank from one year to the next. A commercial variety is a mixed population composed of a number of different strains. These strains may possess a varied degree of response to the factors of environment of a given locality. If a commercial variety is grown for a number of years in one locality its component strains may be expected to have an es- tablished degree of reaction to the environment of that locality. Strains selected out of such a variety will tend to retain their relative rank in respect to a given quantitative character. On the other hand, varieties introduced into a new environment can not be expected in the first years of adaptation to segregate out strains with a fixed type of response to the new environment in regard to their quantitative characters. This brief considera- tion may acount for the different behavior of the local, Aroos- took grown and the Minnesota strains.

BAKING TESTS OF THE PuRE LINES.

The wheat of thirty-one pure lines of the 1918 crop was ground in a small experimental mill in the laboratory of the Rus- sell-Miller Milling Co., Minneapolis, and the flour subjected to

36 Matne AGRICULTURAL EXPERIMENT STATION. 1920

a chemical analysis and a baking test in the laboratory of the Ward Baking Company, New York. The writer is greatly in- debted to Dr. Chas. Hoffman, Chief Chemist of the Ward Bak- ing Company for his careful study of these flours for strength and baking quality.

The results of the chemical analysis of these flours are given in Table 9. The lines are grouped according to their parent va- rieties and their dry gluten rank.

From Table 9 the considerable variations in the gluten con- tent will be first noted. With a number of strains the percent- age of dry gluten is rather high, several of the Aroostook lines exceeding the Minnesota lines in gluten content. Except for a few striking exceptions, there appears to be a relation between the amount and quality of gluten. The Red Fife and Bluestem varieties furnished a large number of.strains with good quality glutens while the Preston, Unnamed and Marquis show a large percentage of strains with fair to poor quality of gluten. The Marquis lines showed uniformly short, stiff glutens of only fair quality. The Minnesota lines with the exception of durum and Marquis furnished strong glutens of good quality.

The data from the baking test with regard to water used, volume of loaf, texture, color of crumb and external appearance of loaf are presented in Table 10. According to Dr. Hoffman’s report the following ingredients were used in each case:

EN Ours eee eae ee Sed 3 300 grams a Viet peer cteenee a ate eiapaae mn Le eit Amount to give correct stiffness Sugaiees natn eu eee 20 grams Sal G8 se cee ie, ae ave re ean eee 444 grams PV eas tee ee a ee ne ee 5 grams

Arkady Yeast Food 1% grams

Temperature set 82° F.

Time of fermentation 4% hours. This is the time when the dough is mixed until it is moulded ready for the pan.

All doughs were well moulded by machinery, but mixed by hand. Baking was done under factory conditions.

on

WHEAT INVESTIGATIONS.

AVAIL, 9)

37

Results of Chemical Analysis of Flours From Pure Lines of

Line No.

2386 2379

2393 2395

Average

2391 2387

2394 2384

Average

2402 2397 2383 2414 2388 2409 2405 2400

Average

2398 2410

Average

2373 2396

Average

Selected from) Variety

Red Fife Red Fife

|Red Fife

Red Fife Red Fife

Bluestem | Bluestem

Preston Preston Preston ‘Preston \Preston Preston Preston

/Preston |

|Marquis

Marquis |

Canada Red Canada Red

Wheat Grown at Aroostook Farm. 1918 Crop. AROOSTOOK LINES. Gluten Ash | Mois- | Wet Dry Ratio | Condition of Gluten per ture per per |Wet to cent per cent cont Dry cent Gluten | 0.70 8.40 37.96 | 13.13 | 2.89 |Soft and sticky—weak gluten 0.60 8.50 36.50 12.74 2.87 Soft elastic gluten with good expanding quality 0.70 8.30 33.55 | 12.18 | 2.75 |Tough gluten of fair | quality 0.67 8.38 382.78 11.78 2.78 Tough, strong gluten 0.66 8.50 33.60 11.78 2.85 ‘Gluten strong and elastic 0.68 8.30 33.70 11.67 2.88 |Good quality gluten, strong and eslastic 0.70 8.50 32.25 11.26 9.89 |Medium quality gluten —somewhat dead 0.67 8.41 34,33 12.08 2.83 0.77 8.40 35.40 12.35 2.87 |Tough, elastic gluten 0.74 8.32 34.55 E25 9.82 |Very good, strong gluten 0.78 8.47 33.85 11.95 9.83 |Strong, tough gluten 0.67 8.57 82.05 11.28 2.84 |Medium softness with | fair quality 0.74 8.44 33.88 11.98 2.84 0.72 8.53 40.38 13.26 3.04 |\Soft and sticky, strength poor 0.66 8.60 39.93 12.84 3.19 Soft and elastic—good quality 0.86 8.70 37.47 12.83 2.92 \Gluten soft and sticky, | fair quality 0.68 8.14 35.05 12.30 2.85 |Elastic gluten of very good quality 0.64 8.20 35.45 12.13 2.92 |\Strong gluten with | very good expand- ing quality 0.79 8.53 34.25 11.85 2.89 |Very soft and sticky 0.79 8.38 32.85 11.23 2.93 |\Soft dead gluten—no elasticity 0.74 | 8.78 31.50 10.89 2.80 Soft sticky, of poor quality 0.74 8.48 35.86 12.17 2.94 | 0.81 | 880 | 31.83 | 11.69 | 2.72 (Stiff, tough gluten— shows fair quality 0.77 8.49 29.20 10.60 2.75 |Short, stiff gluten, | fair strength 0.79 8.65 30.52 11.15 2.74 0.70 8.50 39.21 13.23 2.96 |Soft but elastic and of good quality 0.69 8.20 31.48 10.89 2.91 |Soft and sticky, very low quality 0.69 8.35 35:35 12.02 | 2.94

38 MAINE AGRICULTURAL EXPERIMENT STaTIon. 1920

Results of Chemical Analysis of Flours From Pure Lines of Wheat Grown at Aroostook Farm. 1918 Crop.

—Concluded. Gluten Line Selected from No. Variety Ash | Mois- Wet Dry Ratio Condition of Gluten Z per ture per per Wet to cent per cent cent Dry cent Gluten 2415 Unnamed 0.62 | 840 | 37.40 | 12.22 3.06 Soft and sticky, lacks strength 2370 Unnamed 0.70 8.42 32.65 | 11.53 2.83 Soft, dead gluten— no life ; 2406 Unnamed 0.66 8.56 29.53 10.31 2.86 (Soft, putty-like gluten Average 0.66 8.46 33.19 | 11.35 2.92 MINNESOTA LINES. Gluten Accession Selected from No. Variety | Ash Mois- Wet Dry Ratio Condition of Gluten per ture per per Wet to cent per cent cent Dry cent Gluten 18 Royalton | 0.74 8.20 36.28 12.69 2.8 Very good gluten with (Red) | strong expansive qualities

185 Royalton 0.68 8.28 34.28 | 12.28 | 2.81 |Strong, elastic gluten

(White)

189 Bayne Blue- 0.72 8.30 34.40 11.83 2.91 Tough and strong stem

187 Marquis 0.77 8.30 32.25 11.56 2.79 Short and stiff of

medium quality 182 Speltz Marz 0.90 8.49 27.00- 9.93 2.72 Very dead-like and

(durum) non-elastic Average 0.76 8.31 32.84 11.66 2.82 Average (excluding durum) 0.73 8.25 34.30 12.09 2.86

In Table 10 the pure lines are grouped within their parent varieties according to the volume, expressed in cubic centimeters, of bread loaf produced from their flour. A photograph of each loaf baked from the flour of each strain is given in the accom- panying figures. From these photographs the size, volume, tex- ture and general appearance of each loaf will be noted. A study of the data given in Table 10 will show a number of strains pro- ducing a bread of good volume and very good appearance. . As in the case of other characters the variation in the size of bread loaf is very marked. The Red Fife line No. 2393 produced a

WHEAT INVESTIGATIONS. 39

bread loaf with the highest volume—222I1cc.—of very good tex- ture and good color. Very similar qualities were shown by the Bluestem line 2391 with a loaf volume of 2209 c.c. The third highest loaf volume was produced from flour of another Red Fife strain—No. 2385. The best Minnesota line, Royalton (Red) ranks fourth in volume of loaf. On comparing the data from

TABLE, 10:

Baking Tests of Flours from Pure Lines of Wheat Grown at Aroostook Farm. 1918 Crop.

AROOSTOOK LINES

Line

Selected from

Variety Red Fife Red Fife Red Fife

Red Fife

Red Fife Red Fife Bluestem Bluestem Bluestem

Bluestem

Preston Preston Preston

Preston

205

Loaf Volume in cubic

centi- meters*

2,221

2,153

2,028

2,017

1,903

1,881

2,209

2,096

1,983

1,813 2,068

2,028

1,983

1,892

Texture

| Very good (CHose

Good, close \Very good

|Fair

| Very good Coarse

Very good

Fair |

Poor

Fair

Close

Very close

Very good

Good

White with velvety sheen

Good

Fair

White

Fair

Good

Fair

Good

Verv good—

white

Good

Very good

Good

| External Appearance

Excellent baking qual- ity—produced nice appearing loaf

\Very good size and

appearance. Excellent baking qualities Flour produces good sized loaf, but lacks baking strength Volume good. Has strength and makes good appearing loat Tendency of dough to tear during proofing, causing a rough ap- pearance of loaf Size of loaf fair but flour shows good baking quality Poor baking qualities, appearance of loaf fair

‘Very good appearing

loaf. Flour has good baking strength

Volume good. Flour has good baking qualities

Volume good. Flour shows strength and qualities

Fair size to loaf, fair baking quality

Volume good. Appear- ance good. Baking quality very good

Voluma good. Shows good baking quali- ties

Good volume and ap- pearance. good qual- ity for baking Good appearing loaf. Gluten lacks strength

for good baking re- sults

40 MaINneE AGRICULTURAL EXPERIMENT STATION. 1920

Baking Tests of Flours from Pure Lines of Wheat Grown at Aroostook Farm. 1918 Crop.—Concluded.

Line! Selected | Grams! Loaf

No. from | ,of | Volume Texture Color Variety | water | in cubic of External Appearance | used | centi- Crumb | | | meters* : 2400 Preston 200 1,869 Fair Fair Has poor baking qual- | ities. Lacks strength to give proper ex- pansion ; 2383 Preston 205 1,858 Fair Dark Appearance good. Has ; good baking quali- | ties 2409 Preston | 200 1,858 Good Fair Poor quality for bread ' baking 2405 Preston 200 1,773 Close Dark Fair appearing ‘loaf, but gluten too dead to give proper ex- pansion 2398 Marquis 210 1,949 Good Dark gray Volume good. Flour has strength for baking 2410 |Marquis 210 1,858 Coarse Fair Fair baking quality 23738 |Canada Red 205 2,017 Good Good Volume good _ but

strength too low for good baking results 2396 |Canada Red 205 | 1,745 Coarse Dark ‘Volume poor. Appear- ance of loaf fair. Poor baking quali- : ties 2415 Unnamed 200 | 41,926 Close Fair Appearance of loaf | fair, but gluten too - weak to prevent tearing on the side | of loaf 2370 \Unnamed 200 1,756 Coarse ‘Dark Appearance of loaf good but lacks vol- ums. Quality lacking 2406 Unnamed 199 |} 1,518 Very coarse Dark Very poor quality flour for bread bak- ing

MINNESOTA LINES

| |

A Selected Grams Loat |

= from of Volume Texture Color |

ZB Variety water | in cubic of | External Appearance S6 used centi- Crumb |

LZ meters*

186 Royalton | (Red) 210 2,079 Good Slightly Very good volume and

gray appearance. Very 185 Royalton good quality (White) 210 2,028 Good, close Good Volume and expansion ; P showed good quality 187 Marquis 210 1,903 Very coarse Dark Fair appearance of ; loaf, but flour lacks 180 Haynes strength Bluestem 210 1,881 Close Slightly Size of loaf good, ap- gray pearance good. Bak- ing qualities very 182 Speltz Marz good (durum) 190 1,360 Very coarse Dark No strength to give size and appearance to loaf

*Calculated to 340 grams of flour Der each loaf.

WHEAT INVESTIGATIONS. 41

the baking test with the results of the chemical analysis of the flours it will be noted that the quality of gluten is very well re- flected in the volume and appearance of the bread loaves. With a few rather striking exceptions the volume of loaf appears to follow in a number of cases also the quantity of gluten. This point will be recurred to later.

Since the volume of loaf is at present the most reliable in- dex of flour strength, and further, one of the most important factors in determining the commercial value of bread wheat, it will be of interest to determine how the wheats from the pure lines here considered would rank among the chief American wheats. We may here again refer to the study of several classes of American wheats by L. M. Thomas.” On the basis of ex- tensive baking tests involving 1386 samples Thomas found the - following average volume for each of the five classes of Ameri- can wheats:

S Oiitanevytai towne see Be Re ove ee eine Ja Ds 1,907 c.c. S Ontstippe ts © Clin WANT Le tree eee ee nO dle 1,965 c.c. Dy targeantaernwsln ea tee tees ow let oie ZOVAY) (Ce Idlasealse@cl .. \yatmneies Z~MS) CC: andre di aspiring sete ks Susie eit Tole te 2,421 c.c.

‘Comparing our best line 2393 having a volume of 2,221 c.c. we find that it falls just 200 c.c. short of the average volume of its own class of wheat, the hard spring wheat. From a diagram in Thomas’ bulletin in which the distribution of the 574 samples with regard to volume in the hard spring wheat class are illus- trated we may note that about 15 per cent of hard spring wheat samples had a loaf volume lower than some of ‘our best lines. Further reference to the average loaf volume of American wheats given above, shows that our best lines furnished a loaf volume very considerably higher than the average of the two classes of soft wheat, a higher than the average of the durum class, and equal to the average of the hard red winter class. From this comparison we note that our best lines of wheat are as strong as 15 per cent of the hard red spring wheats and 50 per cent of the hard red winter wheats.

The data given by Thomas are based on samples taken from

the crops of 1908 to 1913 inclusive. If we should consider the.

AIL AS, (ojos. SEIS

——

42 MAINE AGRICULTURAL EXPERIMENT STATION. 1920

strength of wheats determined for one year in a state growing some of the best spring wheats, for instance, North Dakota, we would find that the size of loaf of our best lines approaches still closer the size of the strong wheats grown there. Thus in a report on the baking data for the 1915 crop of North Dakota** wheat we find that the average loaf volume of Bluestem, Fife, Velvet Chaff and Marquis, including all grades is 2307 c.c. In- cluding only the two best paid classes of these four wheats, No. 1 Hard, and No. 1, the average volume of loaf from the flour of these two grades of wheat is only 2271 c.c. or only 50 c.c. above the volume of the best of our pure strains. While these comparisons are not quite fair they nevertheless convey some idea as to the possibilities of growing strong wheats in Northern Maine.

Relative to other points determining the quality of bread, reference to Table 10 will show that the texture and color of crumb with a relatively large proportion of the pure strains was found to be good or very good. Especially the strains of the Red Fife and Preston varieties excel on this point. Most of these breads possessed excellent eating qualities.

RELATION BETWEEN PROTEIN, GLUTEN CONTENT AND SIZE OF Breap LOAF.

With the data on crude protein, dry gluten and loaf vol- ume for these pure lines at hand it is very desirable from the breeding point of view to examine if there is any relation between these three characters which would be of diagnostic value in the breeding work. In Table 11 the pure lines are grouped within their respective parent varieties in the order of their crude protein content with the corresponding rank in glu- ten and loaf volume.

An inspection of Table 11 will reveal an undeniable rela- tionship between the crude protein content, dry gluten and the size of loaf. While this relation is not quite regularly consist- ent it is nevertheless distinct. In regard to the degree of rela- tionship between these three factors the data given in Table 11

*Sanderson, Thomas. The Milling and Baking Data for the 1915 ‘Crop of Wheat. 1917, N. Dakota Bull. No. 132, pp. 61-94.

Relative Rank of the Pure Lines with Respect to Crude Pro-

Accession No.

187 186 185 18) 182

WHEAT INVESTIGATIONS.

TABLE 11

43

tein, Dry Gluten and Loaf Volume. 1918 Crop. AROOSTOOK LINES. Crude Protein Dry Gluten Loaf Volume Variety : per cent | Rank | per cent | Rank c.c. | Rank | | | | | 'Red Fife Hi aE eet ISS | 2d 2 Red Fife BES) |) 13.13 il 2,028 3 Red Fife 12.83 3 12.18 3 2,028 4 Red Fife L220 |e 4: 11.67 4 2,221 1 |Red Fife UPA I By 11.78 5 1,903 6 ‘Red Fife 12.14 6 11.78 6 2,017. 5 |Red Fife 11.74 7 EAC sl ee) 1,881 Ti | Bluestem 13.34 1 12.35 | 1 2,209 il Bluestem 12.65 2 11.95 | 3 2,096 2 |Bluestem 12.48 3 12.25 2 1,983 3 Bluestem 11.91 4 28 |e 1,813 4 |Preston 17) |) al 13.26 | 1 1,892 4 |Preston 13.34 | 2 12.84 2 2,028 2 |Preston 13.22 3 12.83 3 1,858 | 6 |Preston 13.05 4 12.30- | 4 1,983 3 Preston | 12.88 5 12.13 5 2'068) 4) Preston |} 11.91 6 11.85 6 1,858 7 Preston |} Gh aat 7 TBS |) ef 1,773 8 Preston 11.00 8 10.89 8 1,869 | 5 | | ‘Marquis | 12.08 1 THUG) Irae alba TUE" coal Marquis | 11.29 2 10.60 Boa ESR i | ‘Canada Red | 18.40 1 13.28 1 Mammy |) ah Canada Red | 11.86 2 10.80 2 1,745 2 Unnamed | 12.99 | 1 11.53 2 1,756 2 Unnamed | 12.31 | 2 12.22 1 1,926 1 Unnamed ellie || aes 10.31 3 1,518 3 | ea MINNESOTA LINES. | Crude Protein Dry Gluten Loaf Volume Variety | } | | per cent | Rank | per cent| Rank | cc. | Rank | | | Marquis | 13.91 il 11.56 4 1,903 3 ‘Royalton (Red) lpaelseo 2 12.69 1 2,079 1 ‘Royalton (White) | 13.22 8° 12.28 2, 2,028 2 |Haynes Bluestem | 12.94 4 11.83 3 1,881 4 ‘Speltz Marz (durum) | 10.43 5 9.93 5 1,360 5 |

indicate a close relation between the protein and dry gluten. Indeed, the ranks of the crude protein and the dry gluten are completely indentical for the strains of the Preston, Marquis,

44 MAINE AGRICULTURAL EXPERIMENT STATION. 1920

Canada Red, and nearly identical for the strains of Red Fife and Bluestem. In regard to the relation of the two chemical components, protein and gluten, to the baking strength there appears to be, on the whole, some relation between the gluten content and size as well as a still less consistent relation between protein and loaf volume.

These relations are of importance especially as they have been established in this case for pure strains of wheat grown at one limited centre, when the evidence for or against these rela- tions has hitherto been based almost exclusively on analysis of samples from commercial varieties. In this connection it is of special interest to cite the results of two chemists, who worked with materials of an entirely different nature. Shutt?* analyzing flours from wheats representing for the most part pure strains selected and bred pure by Dr. Chas. E. Saunders arrived at the conclusion that “between the protein, gliadin and wet and dry gluten there is a distinct relationship, but there is no evidence of a definite or absolute ratio. The results from both series of flours clearly indicate a distinct relationship between these chemical data (protein, gliadin and gluten) and ‘baking strength’ ——a figure made up chiefly of the values for volume, shape and weight of loaf.” Olson,” on the other hand, working with samples of flours from unidentified varieties and received from mills located in 12 different States concluded that there is no relation between the quality of the flour and the total nitrogen and gluten content. “The volumes of loaves appeared to be inversely proportional to the gluten content.”

DISCUSSION AND CONCLUSIONS.

An analysis of the data here presented brings out the fact that pure strains of wheat isolated from commercial varieties when grown under the same environmental conditions show very distinct differences with respect to the physical and chemi- cal characteristics and the bread making value of their grain. The very small tract of land upon which these pure lines grew,

“Shutt, T. F. The Relationship of Composition to Bread-making value. 1907. Centr. Exp. Farm, Can. Bull. 57, pp. 27-51.

*Olson, G. A. Wheat and Flour Investigations—V, 1917, Wash. Agr. Exp. Sta. Bull. No. 144, pp. 1-86.

WHEAT INVESTIGATIONS. 45

and for which a very considerable degree of homogeneity may safely be assumed, lends particular significance to these differ- ences. For one of these characters, viz, the crude protein con- tent, whose behavior alone could be studied from one year to the next, it was shown that these differences are not mere fluc- tuations but the result of inherent tendencies as evidenced by the coefficient of correlation between the values of one year and those of the next. Of the physical characteristics the kernel weight, the hardness as measured by the percentage of yellow berries, and the color show marked differences. Even in the nursery rows where the strains grew side by side the degree of hardness of the different lines was so marked that out of the original 259 strains 158 were discarded as being soft and opaque. This character while very susceptible to the environment is regarded by Freeman (Loc. cit. 1918) as being controlled by ge- netic factors which determine the greater or less sensitivity of this character to the environment. Leith,?® who also studied the inheritance of the yellow berry, found that while there is no dif- ference between the yellow and hard berry of the same pure line in the production of yellow berry in their progeny, there is a very considerable difference between pure lines in their ten- dency to reproduce hard berries.

No less pronounced are the differences between the indi- vidual lines with respect to the chemical characteristics as will be seen from Tables 5 and 7. The average protein content for all strains in 1917 was higher than in 1918. But in spite of the seasonal variation of the environment affecting the absolute quantity of the protein the fact remains that the individual strains tended to retain their relative protein rank regardless of the seasonal average for this character.

From the close association between protein and gluten it may be inferred that there is a tendency for the pure strains to retain also their gluten rank, though there has been no oppor- tunity yet to study the behavior of this character from year to year.

Perhaps the most pronounced differences between the pure lines are reflected in the quality of their glutens and in the size of bread loaf baked from their flours, as shown in Tables g and 10. While the present data do not convey any informa-

eee eith loc cit:

ee

46 MaAINe AGRICULTURAL EXPERIMENT STATION. 1920

tion as to whether these two characters are heritable, the results secured by Biffen*’, Howard, Leake and Howard?® and Leith??®, furnish substantial evidence showing that strength in wheat is a heritable character retained even under very unfavorable en- vironmental conditions of England and parts of India.

From the data reported in this bulletin it is evident that under the same conditions of environment some strains of wheat will retain a higher degree of hardness, a higher amount and better quality of gluten and produce a larger size of bread loaf than others. The logical deduction from this is that the indiv- iduality of the seed, the inherent characteristics of the wheat strain should enter as an inportant factor into the consideration of the chemical composition and bread making value of wheat.

The present data while covering only a very brief period are nevertheless of practical importance in their bearing upon the possibilities of growing good bread wheats in Northern Maine. Owing to the excellent adaptiveness of Aroostook soil and climate to the potato crop the area devoted to wheat in Aroostook is rather small and the growing of wheat in that sec- tion will be largely a problem of meeting primarily the local con- sumption, in other words, the growing of wheat in Aroostook will be concerned above all in raising good bread wheats for home baking. As to the question of raising good flour for home baking in Aroostook it is believed that our data offer a very satisfactory solution. We quote in this connection from the report of Dr. Hoffman, chief chemist of the Ward Baking Com- pany, upon the baking test of our flours: “A number of the flours showed good quality glutens and made bread of very good quality. For home baking most of these flours are satisfactory and will produce a loaf with excellent eating qualities.”

It is, of course, impossible, as yet, to say that the best pure lines so far selected represent the highest limit of baking strength under Aroostook conditions. Only results from a large number of selected strains tested for several years can give an answer to this question.

Until superior strains best adapted to Aroostook conditions are developed through selection or breeding the practical ques-

"Biffen, R. H. Loc. cit. 1908. *Howard, Albert, Leake, H. M. and Howard, G. L. C. Loc. cit. =Weith, 5:8): 1Z0c. Cit.

WueEat INVESTIGATIONS. 47

tion arises as to whether local or imported wheat should be used for seed. The best policy would be to secure local varieties of wheat of known performance in regard to yield and milling and baking quality. Such wheat varieties, however, are not gener- ally available, and it has been a common practice with the Aroos- took grower to import his wheat seed, usually from the North- west. In importing wheat seed the practice of buying seed from mixed car lots of unknown varieties should be discouraged. Our experience with the Minnesota strains clearly indicates that in order to secure a satisfactory seed it is not enough to import hard wheats from the Northwestern Plains, for certain varieties and strains show a greater capacity for adaptation to humid regions than others. Thus the two strains of the Royalton wheat retained strength under Aroostook conditions yielding grain with a high percentage of protein, very good gluten with strong expansive qualities and good appearing loaves of good volume and very good quality. On the other hand the Marquis strain yielded a gluten with only medium quality and a flour lacking in strength. From this it is evident that an imported variety will first have to be tested for its capacity for retaining its strength under Aroostook conditions, and it is, therefore, essential not only to import seed from known varieties but that these varieties should be as free from admixtures as this is pos- sible with commercial varieties. The more a variety approaches the condition of a pure strain the sooner and the more certain will its degree of adaptability to new conditions be determined.

While no definite recommendations can be made, as yet, as to the best variety for Aroostook conditions, our data furnish some information which may be of practical value. Under Aroos- took conditions the Marquis strains did not make a good show- ing. They all yielded flour with short stiff gluten of only fair quality. Coupled with this rather low quality is a low yielding capacity. This is rather unfortunate as the early maturity of the Marquis would make this variety otherwise very desirable for Aroostook conditions.

The Preston strains are good yielders but only a few excel in quality though a number of them showed a high protein con- tent. The growing of strong strains of the Preston group should be encouraged as they are well adapted to the conditions of Northern Maine.

48 Matne AGRICULTURAL EXPERIMENT STATION. 1920

The Canada Red strains (Ladoga type), while good yield- ers should be dropped on account of their low bread making quality. The Aroostook wheats are frequently mixed to a greater or less degree with these strains which probably lower the quality of the resultant flour.

The Red Fife and Bluestem varieties furnished the strong- est strains yielding flour of very good to excellent baking quality. The Bluestem variety, however, is somewhat later than the Red Fife which is some drawback under Aroostook conditions. Its straw is perhaps not quite so strong as would be desirable and its hairy chaff tends to retain more moisture than the smooth chaffed varieties which may favor the attack by fungi and retard the maturing of the grain.

The Red Fife variety appears to be the best choice. The Red Fife strains yielded the strongest flour and have also a sat- isfactory yielding capacity. The Fife wheats have grown for a number of years in Aroostook County and are well adapted to the conditions of that section. From our experience and the results obtained in England, Canada and Australia it appears that the Red Fife wheats are characterized by a high capacity for adaptation and a strong tendency to retain their strength under unfavorable conditions of environment.

EXPLANATION OF PLATES.

The photographs on plates I, II, and III represent loaves of bread baked from flour of the pure lines of wheat grown at Aroostook Farm. The number under each loaf refers to the line of the wheat strain, except the first five loaves where the figure designates the accession number. Each loaf was baked from 300 grams of flour so that the size of these bread loaves is directly comparable. Note the variation in volume of loaves and texture of crumb.

LG

98T

60F2 : LObC 90K SOre COP

S6EC bOET £6EC 16? 06EC

PLATE 3.

be

et a L

BULLETIN 286

THE VARIATION OF MILK SECRETION WITH AGE. IN JERSEY CATTLE.*

Joun W. GOWEN. SUMMARY

The yield of milk changes definitely with age. The change is a logarithmic change not as commonly supposed a linear change.

The variation of milk yield from cow to cow also changes definitely with age. This change is described by a parabolic function.

The curve for growth is logarithmic. The similarity of the two curves suggests the increase in milk production with. age could be accounted for by growth of the mammry gland.

A large body of exact scientific data on milk production has been accumulated since the time of the establishment of the Advanced Registry system of the Holstein-Friesian Association of America and its later adoption by the associations of the other breeds of dairy cattle. These completed records are unique in several ways, chief among which is the fact that the records of the cows included among them must meet a certain standard of performance. ‘This standard for entry has the effect of cut- ting out certain records of cows of the given breed. Thus the Jersey Registry of Merit say that, “if the test is commenced the day the cow is two years old, or previous to that day, she must produce, within one year from the date the test begins, 250.5 pounds butter-fat. For each day the cow is over two years at the beginning of her year’s test, the amount of butter-fat she must produce in the year is fixed by adding 0.1 (one-tenth) of

*This paper is an abstract of a longer paper on “Studies in Milk Secretion V. On the Variations and Correlations of Milk Secretion with Age in Jersey Cattle by the same author published in Genetics, March 1920. All literature citations should be made to this complete paper.

50 MAINE AGRICULTURAL EXPERIMENT STATION. 1920.

a pound for each such day to the 250.5 pounds required when two years old. This ratio of increase applies until the cow is five years old at the beginning of her test, when the required amount will have reached 360 pounds, which will be the amount of butter-fat required of all cows five years old or over. These standards are based upon one complete year’s record from the time of beginning, regardless of any time which may be lost by being dry or calving during the period.” Three facts are obvi- ously true of the Jersey Registry of Merit cows as compared with a true sample taken at random of the milking cows of the Jersey breed; (1) the cows making up the Registry of Merit are a selected sample; (2) the scale of the selection is linear having its lower limit 250.5 pounds of butter-fat production at 2 years and its upper limit 360 pounds at 5 years and over; (3) this requirement means that for each day of age at test the frequency distributions of years production are cut off perpendicularly at the requirement and only those animals mak- ing greater yields than this are allowed to be entered into the Registry of Merit. The data taken from this Registry of Merit are not the true data for the Jersey breed and conclusions based on it cannot be considered as applying to the breed as a whole or to the general problems of milk secretion.

To supply this need of exact data on the Jersey breed as a whole the Maine Agricultural Experiment Station has obtained the recorded data of one of the largest pure bred Jersey herds known. The data are exceptional in the following ways: (1) The records extend back to the year 1897 when the herd was organized; (2) the animals are practically all straight island stock; (3) they have been under the oversight and direction of one manager since 1901; (4) exact records are kept of .the milk production, butter-fat and butter-fat per cent; (5) many of the individual animals have records for several different jactations. The ‘elimination of variation of the milk production of cows or groups of cows caused by changes in any one or more of these five factors is important for the analysis of the causative mechanism of milk and butter-fat production. It is obvious that these records are free from such variables. They constitute a homogeneous group of data representing the island Jersey under constant conditions of management and climate.

al

Tue VARIATION oF MiLK SECRETION WITH AGE. 51

Such an accumulation of exact statistical data on the cows of the Jersey breed for problems of so much interest, both biologically and economically, warrants the application of ade- quate biometrical methods in their analysis. The necessity of such analysis is now well recognized by most investigators as of the utmost importance to our understanding of the funda- mental principles of physiology which underlie the process of milk secretion itself.

The general problems attacked are those of the individual variation between the lactating functions of different dairy cows at a given age. What is the true type of the frequency distribu- tions of milk production? What relation exists between the mean productions of the successive ages in a true random sam- ple of the Jersey breed? An understanding of these and simi- lar questions is necessary to the fuller utilization of the data found in the herd book of the registry association.

The data used for study are all from normal, healthy cows. Two diseases have been present in the herd, tuberculosis and abortion. The tuberculous animals were all eliminated early in the herd’s history by the use of the tuberculin test. All records from animals which were proven to be tubercular or which aborted were not used. Records from animals normally healthy but sick during a given lactation were not used. All of the

cows have been kept in climatic conditions similar to those of:

Western Virginia.

A word as to the method of keeping the data and its trans- fer to this Station. All records are made at the time of milking on the daily milk sheet for the given cow which are kept in the barn. The milking takes place twice a day, the records are for night and morning. The weekly production taken from these sheets is transferred to the herd ledger by a trained book- keeper. The total production for a given month is found to- gether with the yearly production by adding the weekly totals. All records are recorded to pounds and tenths. The cows are tested bi-monthly by the Babcock test and the percentage of butter-fat is recorded beside its corresponding monthly milk yield. All weighings and readings are recorded immediately after they are made so there is little chance of inaccuracy. From these records the author has extracted 1741 complete 8 months records of healthy cows for milk production. Of these 1741,

52 Maine AGRICULTURAL EXPERIMENT STATION. 1920.

1713 have records for the butter-fat percent. The weighed monthly averages of the bi-monthly test have been used to ob- tain the weighed 8 months average for the 8 months lactation

period chosen for study.

VARIATION OF JERSEY MiLK PropucTION WITH AGE AT CoMm- MENCEMENT OF TEST.

The tabulation of these records in complete eight months records has been done by the author. The chief physical con- stants of the distributions are presented in table 1 together with their probable errors.

TABLE 1.

Constants of Variation of Milk Production for the Successive Ages at Test in Jersey Milk. (8& Months Lactation Period).

Standard | Coefficient Skewness

Deviation | of Variation

——

3 years 0 months 4032.9-431.7, 818.9-422.4

Age at Test Mean |

2 years 0 months to 20.304 .577

3 years 0 months 4 years 0 months 5 years 0 months 6 years 0 months 7 years 0 months 8 years 0 months

to to to to 7 years 0 months to 8 years-0 months to 9 years 0 months

4 years 0 months 4686.5-++46.8) 1101.5+33.1) 5 years 0 months 4992.9-+46.3| 1079.4-+33.0 6 years 0 months 5281.4--57.0| 1262.8+40.3

5536.5--64.5) 1325.3-+-45.6 5314.7-464.9) 1255.02-45.9 5226.4-++77.9) 1302.0--55.1)

012 684 801

23.503-- 21.619 23.911 23.938 23.613 . 24,912-+1.111

865} 907)

+0.2642+-.0548 +0.3532-+.0635 +0.3894-+-.0714 +0.2586.0628

+0.3141+.1173

4938.2-++95.3| 1302.6-++67.4 4838.8-++70.3) 1077.3+49.7 4887.6--20.2) 1249.7-414.3

26.377+1.453 22,.264-++1.028} 25.5694 .310

to 10 years 0 months and above

9 years 0 months 10 years 0 Inonths Total population

The mean milk production for the eight months period, for those cows which are between 2 and 3 years of age, is 4032.9 pounds. From this point the milk production rises rapidly at first then more slowly to a maximum at about 7 years. From this maximum the decline in milk production is less rapid toward the higher ages.

The difference between the production of 2 years and 6 months and the maximum is significant as the difference is about 20 times its probable error (1503.6+71.9). The difference between the maximum production and that for the 10 years and above is ‘also significant as the difference is more than 6 times the probable error (697.7+95.40). This rise and fall of milk

+0.3150-+.0222

THE VARIATION OF MILK SECRETION WITH AGE. 53

production with age has already been shown, by the work of this laboratory, to conform to that group of curves described by a logarithmic function.

The standard deviation varies between 818.9 pounds at the age of 2 to 3 years to 1325.3 pounds at the age of 6 to 7 years. The rise of the standard deviation is more direct than is the rise of the mean curve. From the. maximum the curve remains parallel with the base line for some years when it again drops to a lower deviation at the highest age at test. This rise and fall is significant as judged by the probable errors of the difference. The difference of the maximum standard deviation at 6 to 7 ‘years from that at 2 to 3 years is 506.4 pounds of milk where the probable error is 50.8 pounds or the difference is practically ro times the probable error and the probability that this differ- ence did not come from random sampling is considerably greater than 1,000,000,000 to 1. The difference between the maximum standard deviation and that at Io years and above is 248.0 pounds and the probable error is 67.5 pounds or 3.68 times the probable error and the probability that this difference did not come from random sampling is 75.7 to 1. From these above facts the conclusion seems justified that milk production varies with age at test in such a way that the absolute amount of the variation in production from one cow to another is least at the early age, increases rapidly to a maximum at about 5 years and 6 months, remains at this maximum for about four years then falls again toward the later ages.

The coefficient of variation for the milk production of the successive ages are all high as compared with those already

ALA BILIS, 2

Comparison Data for Coefficients of Variation of Amount of Secretion.

| Coefficient Source of Organism | Secretion of Variation Data Domestic Fowl |Annual Egg Production 34.214.37# | Pearl (Barred Plymouth Rock) | Cattle British Holstein \Gallons of milk 25.72.37 Gavin Cattle Ayrshire 'Gallons of milk 24.18+.50* Vigor Cattle Jerseys /Pounds of milk 25.56.31 This paper

# Probable error calculated by author from total given in paper. Me cated by the author from the means and standard deviation as given by T. Vigor. } Bre

54 MatIneé AGRICULTURAL EXPERIMENT STATION. 1920.

known for other similar data. This large size in the coefficients is brought out by table 2.

Of the substance studied annual egg production is the most variable. The coefficient of variation for milk production cor- responds remarkably well considering the diversity of the sources from which they are taken. They all go to show that milk production varies around a mean of 25 or is about g per cent less variable than egg production. This difference is sig- nificant. The mechanism of the secretion of the sum of the egg parts to form the egg is shown to be more variable than that for the secretion of the milk parts to form the milk. Taking this reasoning back to its ultimate source, such a significant difference shows that the secretory cells of the mammary glands work with greater precision than do the cells of the oviduct. Such a difference in the variation of action of the two sets of cells would seem to indicate a greater approach to perfection, in a mechanical sense, in the cells of the udder than is true of the cells of the oviduct.

The skewness of the two sets of data furnish another in- teresting contrast. The skewness of annual egg production for Barred Plymouth Rock hens is —o.205, whereas the skewness for Jersey milk production is +0.315. Not only is the sign different but the actual amount is different. This cannot be explained on the basis of any selection for high producers that may have taken place as both sets of data are about equally sub- ject to such selection. The data for the successive age groups all goes to show that where the distributions for milk production are skew they are all plus. This lends further strength to the belief that the distribution for the yearly production of the two sets of glands is skew in opposite directions. One thing is com- mon in the two cases, the skewness in each is small in amount. This is of especial importance as it shows the approach to the value where the typical Gaussian curve of error may describe these functions.

This question is of great importance in considering the milking records of advanced registry animals where it is neces- sary to form an opinion of the capacities of a breed from the milking abilities of a selected sample where the selection is not at random but removes those cows producing under a certain amount.

THE VARIATION OF MILK SECRETION WITH AGE. 55

This feature of the problem is seen in the work of Reitz (1909) on the inheritance of butter-fat production in Holstein- Friesian cattle. The data for this problem came from the ad- vanced registry of Holstein Friesian cattle. Since as a require- ment to entry in this registry the cows must produce more than a certain amount of butter-fat, the correlation from such data measuring the strength of inheritance are subject to a double selection. Rietz in correcting for this selection uses the method devised by Pearson to determine the whole of a normal curve when a portion of it is known. The accuracy of the corrections depends then on the curves for milk production and for butter- fat percentage being normal.

When the frequency curves for milk production of these data are anlayzed by the method of moments for their type equations it is found that five of the curves belong to the type I group, four to the type II group and one to the type 1V group. Nine of the ten curves are of limited range. These limits are of interest as they indicate the maximum and minimum produc- tions to be expected from this group of cows. The limits of the range of the type II curves are on the whole somewhat un- derestimated. The type I curves more nearly record the facts of the case save in the curve for the total population where the range is considerably overestimated due to the slow approach to zero of the ends of the distribution. For such a distribution it must be remembered that the frequency becomes very small as the ordinary limits of the observation are passed.

On the whole the fit of the curves is excellent comparing favorably with that of any similar data. It is evident therefore that the skew frequency curves describe these data well. The question now arises in view of the difficulty of correcting for this skewness in correlation studies on the heredity of milk pro- duction, if the fit of the Gaussian curves is sufficiently good to allow of the use of this method.

When these curves are used it is found that the normal curves do not describe milk production at the different ages or for the total curve as well as do the proper type curves. The difference in the fit of the type curves over the normal curve comes especially in the description of the tails and the skewness of these distributions. The fit of the normal curves is not bad, however, and in all probability no serious error would be made

56

in the use of them for the determination of the corrections to be applied to the correlations from double-selected data as was

Marne AGRICULTURAL EXPERIMENT STATION.

done by Rietz.

1920.

THE CORRELATION OF JERSEY 8 MontHs MiILx PrRopUCcTION

WITH THE AGE.AT TEST.

Common knowledge among dairymen is that milk produc- tion and age at the commencement of the lactation are corre- lated in such a way that advancing age means increased produc-

Age at Test for

8000— 8500

* 8500— 9000 9500—10000

~ 9000— 9500 10000-—10500

10500-11000

i) ad jal e

HOD OOO et ot -e

tion. The opinion is general that this increase is a linear one. TABLE 3. Correlation Surface for Pounds of Milk and Jersey Cattle. (Lactation Period 8 Months). She 18/3) 22) 21s) aie les Sanaa iS cS iS 7 > = iS S 2 | S j Ta Ss 5 = rin a oD of Sl a) VT) IS fs cS i ~~. | @ esigieisielelslislel| slalteleles 1:6— 2:0 25) sis} oe2 arch TA =e) | 4| 19) 54) 51| 56 42| 19] 7| 2 lcaedl 2:6— 3:0 i} 4) 6°71) 12) 4| 4 al 3:0— 3:6 4| 21| 22} 33) 30} 20) 13] 9 3 2] -| 3:6— 4:0 3| 3} 6| 9} 14| 19] 16) -8| 7 31 5] 2 4:0 4:6 |) 2) She a5| Si). (24) 325) (sable 7 |e shiiet ane 4:6— 5:0 7| 11] 13| 20] 24] 13] 8 3] 5] 3 5:0— 5:6 7+ 13| 16] 14| 231 10| 5] 4 9) .2 5:6— 6:0 5| 13] 12] 19} 17] 15] 14] 9 4| (8 6:0— 6:6 2} 2) Io} 14) 8} 14} 19] 12] -5] 6] 3 6:6— 7:0 4) 6| 9). 15] 15| 12] 16° 8| 7 4 7:0— 7:6 2} 1] 4| 5] Je 1) 19] 7] 6| 5 5]- 2 7:6— 8:0 3} 11] “S}) <9) (9) 14) a4) 9) 5) a 8:0— 8:6 A}. 2) a] a] 10|) 15] 22s), er ee ott 8:6— 9:0 2) St! 8 <8] 12 "ait eal tae 9:0— 9:6 Ne) ea ee Al et i 3} (5) eal 9:6-10:0 1 3} i} “at 4l 6) 8) Toit 5] “elo es 10:0—-10:6 | I} Sp 59) Jel ale Ould 4 eae 10:6-11:0 1 3) 53) sal “al Sa) Vel ster erie 11:0-11:6 D3) 2h) Ge Sh Spa aa 11:6—-12:0 20h ae ea a "12:0-12:6 tl 3 Peed aes big | 12:6-13:0 1 Wii ae i 4 ome | 13:0-13:6 1 rif pa dd 3:6-14:0 1 heal 14:0-14:6 ul | 14:6-15:0 15:0-15:6 1 | 15:6-16:0 | | | Total 1 13, 50 166) 212 279 274| 266) 162 127, 82) 58| 26

B

¥ 4,

Tue VarIATION oF Mitx SEcRETION wiTH AGE. 57

This opinion has been shown to be erroneous by previous work of this laboratory on the statistics of the 7 day records of Ameri- can Jersey cattle found in “Jersey: sires with their tested daugh- ters,” published by the American Jersey Cattle Club. In this work seven day milk production is shown to be a logarithmic curve and not a straight line.

It remains to be shown that this same type of curve des- cribes a true random sample of the Jersey breed for a longer milk period. Data of this sort are important for several reasons chief among which, both practically and theoretically, is the necessity of having suitable correction factors for age to allow “comparison of milk records at different periods in the lives of different cows. Toward the solution of this problem the fol- lowing facts are necessary; what correlation exists between age and milk production; is this correlation sufficient so that it must be taken into account in considering records of different cows at different ages; what is the equation of the regression line between these two variables. Table 3 furnishes the data neces- sary for this study.

The correlation and its accompanying constants for these two variables are shown in Table 4.

TABLE 4.

Constants Measuring the Association between Amount of Eight Months Milk Produced and Age at Test of Jersey Cows.

E 7 | aan Or ee

0.2596-.0151 | 0.4283-+.0132 | 0.1689-+.0201 | 0.1161.0108

This table makes clear several facts concerning the influence of age on milk production. The correlation of +0.2596+.0151 shows that age at test and milk production are significantly correlated variates. Taken in conjunction with the correlation ratio it shows clearly that age of the cow at commencement of test must be considered in comparing the records of different cows if the conclusion from the comparison is to be valid. The value of the correlation ratio +0.4283-+.0132 is considerably

58 Maine AcricuLTuRAL EXPERIMENT Station. 1920.

higher than the correlation coefficient. This difference is shown to be highly significant by the value of »—r -+0.1689=.0201. It is altogether probable therefore that the regression of age on milk production is a skew regression. This is shown to be a fact by the constant to measure such skewness. 4? —r* O.1161+.0108 is about II times its probable error. The re- gression is therefore known to be skew. Since this is true the correlation ratio is a better measure of the true correlation than is the correlation. coefficient. The relation of age at test is then doubly significant in any comparison of the records of two cows. The regression having been shown to be skew it

becomes necessary to deal with it separately.

TYPE OF THE REGRESSION OF MILK PRODUCTION ON AGE OF JERSEY CATTLE.

The means for each array of age have been calculated.

From these means the theoretical curve conforming to the general logarithmic type has been calculated by the method of least squares. The equation to this curve is

==3 387.91 2—99.883.1—.4874°+-2896.219 Log «x

where « is taken in six months intervals from an origin at I year and 3 months.

The observations at the higher ages vary a good deal as they are based on small numbers. The theoretical curve strikes through them quite accurately when the unevenness of the ob- served curve is considered. When we calculate the x? by the method of Slutsky we find that 5 observations contribute a sum of 28.80 to the total of 45.41. These observations are at ages 2 years 9 months, 3 years 3 months, 6 years 9 months, 7 years g months and g years 3 months. If we measure the fit by the total x* 45.41 it is poor. Considering the above mentioned five observations in connection with the other observations it is seen that two of them are plus and three are minus quantities. Not only that but they come at places in the curve so that they would practically counteract each other if the first smoothed curve were used as the observational. It seems altogether reasonable

THE VARIATION OF MILK SECRETION WITH AGE. 59

therefore to consider the fit of this curve measured by a_ ? somewhat more than 17.00 or what would correspond to the P of.a very good fit.

The equation of the curve has many practical uses aside from its interest in a scientific sense. By its use the records of cows at different ages may be brought to the same basis for comparison whether it be for milk inheritance studies, analysis of judging experiments or the like. The time of the theoretical maximum of milk production may be easily calculated from it by differentiation. This maximum is shown to be 7 years 2.4 months a figure considerably above the age customarily called mature form. Further the curve shows that the method used in advanced registry work of determining the amount a Jersey cow should produce for the Register of Merit is falacious in that it is a linear method and does not recognize this logarithmic nature of milk production. In a previous paper the average fat per cent of Jersey cows is given as 5.12. Assuming this figure and dividing the pounds of butter-fat by it gives us the average requirement for milk production in one year. Supposing that 34 of the year’s records is made in the first 8 months of lacta- tion (a figure reasonably close to the expected (Peark 1915) the required production is found to be 3600 pounds at 2 years and 5200 pounds at 5 years.

Causally considered the logarithmic nature of milk produc- tion is of a good deal of interest. The work of a number of students of growth, beginning with Minot’s notable studies on rabbits have shown that the phenomena of growth is also a logarithmic function of age. This law appears of wide general application as the work of Lewenz and Pearson have shown it holds for growth in children; Donaldson, Hatai and Jackson have shown it is of general application to the growth of certain organs in the white rat and Pearl and Surface have shown it true for ceratophyllum and corn. It seems, therefore, altogether likely that the mammary glands of the cow also follows this rule. Should this prove true the increase of milk production with age seems of much significance in paralleling these growth phenomena. This paralleling of the two functions would, in fact, seem to indicate causal relation between the two in that the increase in milk production may depend chiefly on the in- crease in actual mass of the mammary gland due to growth of

60 Maine AGRICULTURAL EXPERIMENT StTATIon. 1920.

this organ and not due to any relative increase in the ability of the cells to secrete more milk.

In a Bulletin to follow this, the relation of the milk yield for one lactation in comparison with that of another subsequent lactation will be analyzed for the same data as presented here.

BULLETIN 287

SELF STERILITY AND CROSS STERILITY IN THE APPLE.!

Joun W. Gowen.

SUMMARY

The results herein presented show that every apple grower should provide suitable varieties for pollinators if- large de- pendable crops are to be secured.

The results presented in Tables 1 and 2 show the apple varieties which will self fertilize. No difference is noted in the fruit set when a variety is self pollinated, when it is pollinated with the pollen from different flowers on the same tree, or when it is pollinated with pollen from different trees of the same variety.

A large amount of sterility is observed in the different varieties. Out of 119 varieties only 42 set fruit, and of that 42 only 15 had a set of fruit which was even moderately commer- cially profitable.

Tables 3 and 4 show the results of cross pollinations with- in the apple. Most varieties are capable of ready cross fertili- zation with the pollen of other varieties. Over 34 of those varieties pollinated with pollen of other varieties set fruit sat- isfactorily.

Results are presented to show that it is necessary to test a variety for cross compatibility before any conclusion can be drawn for the variety.

As pointed out the yield of orchards made up of one block of self sterile trees may be materially increased by the intro- duction of other varieties.

The size, color, and quality of the fruit is shown to remain practically the same as the standard for the mother parent.

‘Papers from the Biological Laboratory, Maine Agricultural Experi. ment Station, No. 133. i

62 MAINE AGRICULTURAL EXPERIMENT STATION. 192U.

The number of good seeds in the crossed apples is greater than in those which are selfed.

The causes of self sterility in the apple are external and internal. The external, weather, spraying, insects, and disease, are somewhat within the control of the grower.

The chief internal cause of sterility is the slowness of growth of the pollen tube in the selfed style as against that in the crossed style.

Aside from the environmental factors, weather conditions at the blooming period, etc., there is an inborn tendency of certain plants not to produce fruit when fertilized by their own pollen or the pollen of certain varieties within their own species or different species. Among the plants with a well marked tendency in this direction of self sterility and cross sterility is the apple. The tendency of certain of the more common varie- ties of this species is apparently quite distinct and well marked, within other varieties the trees seem to self fertilize readily with their own pollen. It is of especial importance to the prac- tical grower here in Maine to know what varieties are self fer- tile and what varieties should have other varieties near by so that the necessary crossing may take place. It is further of importance to know what varieties of those that must be crossed to produce a fair yield, should be planted together so that the best yield and quality of fruit may be obtained. A large amount of time has been devoted to the solution of this problem. by the staff of the Biological Laboratory of the Maine Station.

MATERIALS AND METHODS.

The apple orchards and scattering apple trees of Highmoor Farm total to approximately 3000 trees. When the grafts are included there are about 25 different varieties represented with- in this group of 3000 trees. The experiments herein described include 16 of these varieties. Controlled crosses have been made between these varieties. The apples resulting from these. crosses were measured. The number of good seed and the number of poor seed were determined for each cross. The ger- mination of these seeds when planted out doors in a cold frame

Se_F STERILITY AND Cross STERILITY IN THE APPLE. 63

was recorded in connection with the data on transplantation. These data all bear on the problem of self sterility and cross sterility in the apple and will be used in connection with this study. The publication of the results obtained from the crosses, the bearing ability of the seedling trees, and the quality of the resulting apples will form the basis of other reports on the orchard work of the Biological Laboratory.

The sterility tests are made in four ways. To test for self sterility the unopen buds are inclosed in a ten pound paper bag. These bagged flowers are treated in two ways; (a) the bags _ are left undisturbed until the fruit is set; (b) the bags are opened at the height of the bloom and the pollen from the an- thers brushed over onto the stigmas, the bags replaced and left until the fruit is set.

The tests for cross sterility are likewise made in two ways; sterility between members of the same variety and sterility be- tween different varieties. All of this was done with emascu- lated flowers, the pollen transfers being made with camel’s hair brushes. In each case the flowers, both emasculated and pol- linated were covered with paper bags, care being used in the removal for pollination and subsequent replacing of the paper bag to prevent accidental pollination.

When the fruit is set the paper bags used in the pollination work are replaced with cheese cloth bags. All the crosses made are tagged with a distinctive number to prevent any pedigree errors.

SELF STERILITY AND SELF FERTILITY.

In table 1 are shown the result of the crosses involving the pollen from a flower cluster being placed on the pistils of that same flower cluster or a different flower cluster of the same trees or different trees. The flower clusters which are only bagged depend, of course, on chance agencies to transport the pollen from the anthers to the stigmas. Those flowers which have the pollen transferred from the anthers of the flower clus- ter to the stigmas of the same flower cluster by means of the camel’s hair brush brushing the pollen across from the one to _ the other eliminate this chance element. ‘The average number of flowers worked to each flower cluster was six. The results

64 MAINE AGRICULTURAL EXPERIMENT STATION. 1920.

as given in table I are all for clusters which did or did not de- velop fruit. If it is desired to determine the fruit which set per flower the results should be multiplied by this number to obtain the number of fruit buds worked.

The selfings which matured apples are the only ones which are recorded as successful. Many of those which fall in the unsuccessful group did start to develop and some even remained after the June drop for a short time. These are not recorded, however, since this paper deals with this problem chiefly from the viewpoint of the mature, marketable fruit.

TABLE 1.

Fertility of the Ovule to Pollen Within the Same Variety. METHOD OF POLLEN APPLICATION.

‘Pollinated with the Flower cluster bagged Pollinated with pol- pollen of a different

=These apples were very poor specimens from which no seeds germinated. The seeds themselves were shrunken and shriveled.

From this table it is clear that most varieties of apples show more or less pronounced self sterility. Within the twelve varie- ties under consideration only four showed any fertility to their own pollen. For those which showed such fertility the Wealthy was self fertile once, the Duchess was doubtfully self fertile in one out of four trees; the Baldwin was self fertile in five out of forty crosses and the Northern Spy was doubtfully ‘self fer- tile in one out of thirty-nine trials. It is clear from these results that the proportion of the flowers which are self fertile to their

Lib Si}

Setr STERILITY AND Cross STERILITY IN THE APPLE. 305

own pollen is slight even with those varieties which will self fertilize. This is especially true when it is realized that each of the selfings within table 1 represent the flower cluster and not individual flowers.

The results from the different methods of pollination are chiefly negative in character. The three different groups show no material difference in the set of the fruit for the three meth- ods. This is of interest in connection with the results of polli- nation with pollen of the same tree and the results of pollina- tion with pollen of a different tree but of the same variety. The results are in each case approximately the same. This would be expected in view of the probable fact that the trees of a given ~ vatiety are ultimately of the same origin, coming as they do from the same original seedling. Such results indicate the rela- tive stability of the buds and the trees which grow from them in their presumably hereditary behavior to crossing with dii- ferent kinds of pollen.

It shows further the probability that the planting of a large block of trees of the same variety, if it is self sterile, will not tend to a larger crop of fruit because for these self sterile varie- ties the pollen of other trees of the same variety is no more compatible than the pollen of the tree itself when applied to the stigmas. |

It is of considerable interest to gather together the results on the self sterility of the apples varieties as it has been deter- mined by the different states, both to determine on as large numbers as possible the amount of sterility which exists and ' also to see whether the technique or climatic conditions of one state favor the fruiting of varieties normally incompatible to their own pollen in other climates. For this purpose the results on self sterility of the different varieties have been collected and brought together in table 2.

TABLE 2.

Self Fertility and Self Sterility m the Varieties of the Apple.

] 6 : | Number selfed | Number fruit Number fruit not Variety matured matured Arkansas Black+ | 100 100 - Autumn Sweet* ie 50 : 50 Baldwint 169 1 168

66 MAINE AGRICULTURAL EXPERIMENT STATION.

1920.

Self Fertility and Self Sterility in the Varieties of the Apple.

Variety

Baldwin Baldwin‘

Bailey’s Sweet* Ben Davis+

Ben Davis1?

Ben Davis

Ben Davis Bethlehemite* Bietigheimer* Bellflower (Yellow)¢# Bottle Greening+ Bough, Sweet? Canada Red+ Canada Reinette* Canada Sweet* Colvert*

Crab

Jelaware* Domine+

Duchess

Dutch Mignonne* Early Harvest Early Harvest? Early Ripe? Early Strawberry* English Russett? Esopus (Spitzenburg)? Ewalt+*

Fallwine* Fallawater*

Fall Jenneting* Fameuset Fanny?

Gano+*

Gilpin (Carthouse)? Golden Russett Golden Sweet* Gravenstein2 Gravenstein* Great Bearer* Green Sweet* Grimes Golden* Grimes12

Grimes?

Haas*

Hanwell Souring* Hawley

Holland Beauty* Holland Pippin+ Hoover’s Red+ Hurlbert Sweet Hydes Keeper+ Jonathan13 Jonathan* Jewett’s Red+ July, Fourth of? King

King of Tompkins Co.4 Keswick Codlin* Longfellow* Limbertwig*

Lily of Kent? Lily of Kent? Maiden’s Blush* Mann+

Mammoth Black Twig* May*

McMahon White*

—Continued.

Number _ selfed

Number fruit matured

1?

moo fe

a

14 37

Number fruit not matured

SELF STERILITY AND Cross STERILITY IN THE APPLE. 07

Self Fertility and Self Sterility in the Varieties of the Apple.

—Continued. Number selfed Number fruit Number fruit not Variety matured matured .

McIntosh Red 28 28 Melon* 50 50 Melon Sweet‘ 50 50 Missouri Pippin* 50 50 Missouri Pippin? ’ 57 57 Missouri Pippin? 150 150 Montreal Beauty (crab)+ 100 100 Munson Sweet+ 50 50 Nero? 150 150 Newtown 100 66 $4 Northern Spy 38 1? 37 Northern Spyt 19 19 Northern Sweet? 113 113 Oldenburg* 100 5 95 Ortley+ 100 100 Paradise Sweet* 100 100 Paragon? 195 195 Paragon? i 180 180 Pewaukee* 50 50 Portert 52 52 Pryor’s Red‘ 50 2 43 Pumkin Russett* 100 16 84 Astrachan2 200 12 188 Ralls* OORee 100 Rambo 100 2 98 Red Astrachan 4 4 Red Astrachant 16 16 Red Canadat 80 3 80 Red Cheek Pippin‘ 100 100 Red Golden Pippin¢ 50 50 Rhode Island Greening 14 14 Rhode Island Greening* 100 100 Rhode Island Greening? 703 703 Romanite* 100 100 Rome Beauty 100 100 Roseaut 120 120 Roxbury (Russett)1 119 119 Salome* 100 100 Scott’s Winter+ 100 39 61 Shiawassee* 100 i 23 77 Spitzenburg+ 100 if 93 Stark¢ 100 1 99 Stark? 150 150 Stayman3’ 161 161 Stayman? 106 106 Steel’s Red* 50 50 Strawberry2 200 1 | 199 Red Streak2 200 1 199 St. Lawrence‘ 100 100 Summer Permain‘ 50 50 Summer Queen‘ 100 100 Sweet Bough* 50

Tolman (Sweet)? 223 223 Tolman Sweet+ 100 100 Transcendent Crab‘ 100 100 Trumble Sweet+ 100 100 Twenty Ounce* , 100 100 Wealthy 1 1

Wagener? 50 3 47 Washington* 50 7 43 Wealthy 28 28 Wealthy+ 50 50 Westfield (Seek-no-further)1 485 485 Western Beauty+ 50 50 Williams (Favorite)1 63 63 Williams Favorite? 150 150 Willow Twig* 50 2 48 Winesap+ 100 100

’

68 MatneE AGRICULTURAL EXPERIMENT STATION. 1920.

Self Fertility and Self Sterility in the Varieties of the Apple.

—Concluded. Number selied Number fruit Number fruit not Variety ; matured matured Winesap? 309 300 Winesap?3 550 z 548 White Pippin* 100 26 74 Whitney's Crab+ 100 4 96 Yellow Transparent? 363 20 3 Yellow Transparent+ 25 2 : 23 York Imperial* 100 100 York Imperial? 134 1? 133

Even a cursory examination of this table will show that the degree of self fertility in the apple is quite generally insig- nificant. Within this group of one hundred and nineteen varie- ties only 42 or less than half are known to have self-fertilized and set fruit. Of these 42 varieties only 15 set fruit in any numbers, the rest had only one or two fruit which matured rep- resenting something less than five per cent of the total number of crosses made.

Table 2 shows one of the best commercial varieties, the Baldwin to be self fertile in Maine and elsewhere. Of the other leading commercial varieties Rhode Island Greening, Golden Russett, Tolman Sweet, Twenty Ounce, McIntosh and Graven- stein proved to be self sterile in all tests. The varieties North- ern Spy, Esopus Spitzenburg, Ben Davis, Fameuse and Olden- burg proved very slightly fertile. Of the other commercial varieties which proved somewhat more fertile might be men- tioned the Jonathan, Early Harvest and Yellow Transparent.

Considerable difference is evidenced by the record of the set of fruit of a variety within the different states. The Bald- win sets a very limited number of fruit in Vermont whereas in Maine and Oregon its set of fruit was more numerous. The Ben Davis in Maine and Vermont set no fruit whereas in Arkansas and Oregon it set a limited number of apples. The Red Astrachan proved self sterile in Maine and Vermont but with a test made in Maryland set fruit on self fertilization. These results make it seem probable that the environmental conditions of the different states affect the self fertility of these differently. Caution is consequently necessary in applying the results of one state to that of another.

Sed CT EN ee A ee, ae ee hy ye

Tee See ee ON Pee ee eee

Pg eee

Se

Setr STERILITY AND Cross STERILITY IN THE APPLE. 69

Cross FERTILITY AND Cross STERILITY IN THE APPLE.

In table 3 are shown the results of crossing one variety with the pollen of another variety. The first column records the female variety and the second column the pollen variety. In the column marked “Successful pollination” are recorded the number of pollinations which produced mature apples. In the column “Unsuccessful pollination” are recorded the num-

ber of flower clusters emasculated and pollinated.

TABLE 3.

Cross Fertility and Cross Sterility in the Apple.

*Not found until sueeeeding year when seeds were no good.

+Had six shrivelled seeds.

Female Parent Pollen Parent Successful Unsuccessful Pollination Pollination Ben Davis Baldwin ik? 2b] Duchess Baldwin 20 Golden Russett Baldwin 2 Northern Spy Baldwin lt 12 Red Astrachan Baldwin 4 Rhode Island Greening Baldwin 2 Baldwin Ben Davis 4 Golden Russett Ben Davis 5 9 Hurlbert Sweet Ben Davis 3 q McIntosh Red Ben Davis 5 Northern Spy Ben Davis 2 Rhode Island Greening Ben Davis 2 Ben Davis Canada Red 5 18 Hurlbert Sweet Canada Red 1 1 Ben Davis Crab 9 4 Early Harvest Crab 7 Baldwin Duchess 1 9 Early Harvest Duchess 1 al Red Astrachan Duchess 1 1 Baldwin Golden Russett 22 5 Ben Davis Golden Russett 5 40 Northern Spy Golden Russstt 2 Red Astrachan Golden Russstt 2 Rhode Island Greening Golden Russett 2 Ben Davis Gravenstein 3 22 Ben Davis Hurlbert Sweet 1i Early Harvest Hurlbert Sweet 2 Ben Davis MeIntosh Red 40 19: - Duchess McIntosh Red ‘ 4, Early Harvest McIntosh Red 2: Hurlbert Sweet MelIntosh Red 4 Baldwin Northern Spy 2: Ben Davis Northern Spy 10 10 Golden Russett Northern Spy 3 23: Ben Davis Opalescent 19 23 Hurlbert Sweet Opalescsnt 5. McIntosh Red Opaleseent 2 iy Baldwin Rhode Island Greening 2 Ben Davis Rhode Island Greening 2 Golden Russett Rhode Island Greening 2: Northern Spy Rhode Island Greening 2 Ben Davis St. Lawrence 3° 9: Ben Davis Wealthy 9. By

70 MAINE AGRICULTURAL EXPERIMENT STATION. 1920.

Forty-three different kinds of crosses were tried in testing for any cross sterility which might exist between the different varieties. Of these crosses 20 proved compatible and formed fruit. Only two of the crosses tried more than to times failed to set fruit. These two crosses were Duchess female x Baldwin pollen and Ben Davis female x Hurlbert Sweet pollen. When the cross was made the other way Baldwin female x Duchess pollen and Hurlbert Sweet female x Ben Davis pollen the cross was successful and fruit was matured. It is desirable, there- fore, to leave those crosses which did not set fruit in abeyance until such time as more data can be collected for them before any definite conclusion is drawn on their cross sterility under Maine conditions.

Of those trees which proved fertile certain varieties stand out as quite desirable for commercial plantings. Considering the number of crosses made in conjunction with the amount of fruit set Ben Davis pollen proved quite successful with Golden Russett female; Golden Russett pollen proved to set a high percentage of the fruit when crossed with the Baldwin; Golden Russett pollen crossed fairly well with the Ben Davis; McIn- tosh Red pollen proved very desirable for crossing on Ben Davis female. The same was also true for the pollen of Northern Spy, Opalescent, Crab and Wealthy when crossed with Ben Davis.

Table 4 gives the same data for the varieties which have been tested for cross fertility as that given in table 2 for the self sterile varieties. The data are presented for those which are compatible and set fruit on crossing and those which did not prove compatible and did not form fruit. The crosses which are marked plus (++) or yes proved to set fruit on cross- ing. Those marked minus (—) did not set fruit. After those which did not set fruit is given the number of trials that were made for the given cross. From these data some estimate may be made of the probability that fruit might be set on a further crossing of these same varieties.

The percentage of fruit set or the degree of compatibility of the cross is indicated where it is known by the number of plus signs. The + sign represents a very low percentage of fruit set with only a few number of trials. The + sign shows that a low percentage of fruit was set, the number of trials be-

SELF STERILITY AND Cross STERILITY IN THE APPLE. 7\

ing large. The +-+ sign shows a greater percentage of fruit set. The +-++-+ sign indicates a cross which proved highly compatible by the percentage of fruit which resulted from the cross.

As these data represent the crosses which have been made in several states it gives an opportunity to compare the fruit set of the same cross under the different environmental con- ditions.

TABLE 4.

Cross Fertility in the Apple.

Variety Pollen Fruit Set No. of Trials

Arkansas Black14 x Jonathan i yes

Baldwin x Ben Davis —_— 4 Baldwin x Duchess +

Baldwin x Golden Russett SP Spa

Baldwin x Northern Spy — 2 Baldwin x Rhode Island Greening — 2 Ben Davis x Baldwin +

Ben Davis x Canada Red +

Ben Davis x Crab tot

Ben Davis® x Esopus yes

Ben Davis x Golden Russett +

Ben Davis x Gravenstein ° +

Ben Davis® x. Green Newton yes

Ben Davis18 x Grimes ++

Ben Davis x Hurlbert Sweet —_ 11 Ben Davis18 x Jonathan +

Ben Davis® x Jonathan yes

Ben Davis14 x Jonathan yes

Ben Davis® x McIntosh : yes

Ben Davis x McIntosh Red +++

Ben Davis® x Mother yes

Ben Davis14 x Newtown yes

Ben Davis x .Northern Spy ++

Ben Davis x Opalescent ++

Ben Davis x Rhode Island Greening —_ 2 Ben Davis14 x Rome yes

Ben Davis14 x Spitzenburg yes

Ben Davis x St. Lawrence +

Ben Davis14 x Wagener yes

Ben Davis x Wealthy Stneete

Ben Davis18 x Winesap ais

Black Ben Davis11 x Hydes Keeper H yes

Black Ben Davis11 x Willow Twig | yes

Blenhein Orange?! x Hanwell Souring yes

Blenhein Orange?! x Arkansas Black yes

Bienhein Orange? x Jonathan yes

Bloomfield’ x Delicious +++

Bloomfield’ x Oldenburg +

Bottle Greening?! x Pewaukee yes

Bottle Greening11 x Charlottenthaler yes

Delicious’ x Grimes _— 64 Delicious!4 x Jonathan yes

Duchess x Baldwin _ 20 Duchess x McIntosh Red _ 4 Early Harvest x Crab — 7 Early Harvest x Duchess ap

Early Harvest8 x Early Ripe SE

Early Harvest x Hurlbert Sweet — 2 Early Harvest x McIntosh Red — 2

72 MAINE AGRICULTURAL EXPERIMENT STATION.

Cross Fertility in the Apple—Continued.

Variety

Early Harvest§ Early Harvest® Early Harvest§ Early Ripe§ Early Ripe® Early Ripe® Early Ripe® Early Ripe® Early Ripe? Early Ripe® Early Ripe§ Early Ripe§ Early Ripe?

Golden Russett Golden Russett Golden Russett Golden Russett Gravenstein$® Gravenstein1+ Gravenstein1+ Grimes® Grimest? Grimes® Grimes13

Grimes Golden1t Grimes§ Grimes13 Hanwell Souring+1 Hanwell Souring11t Hoover’s Red11 Hoover’s Redit Hoover’s Red1t Hurlbert Sweet Hurlbert Sweet Hurlbert Sweet Hurlbert Sweet Hyde’s Keeperit Ingram’ Ingram§s& Jonathan!* Jonathanit Jonathan!* Jonathan: Jonathan13 Jonathant+ Jonathan?1 Jonathan!+ Jonathan1+ Jonathan? Jonathan1+ Jonathan? Keswick Codlin1t Keswick Codlinit Lily of Kent® Limbertwig1t Limbertwig11 Maiden’s Blush11t

Mammoth Black Twig1? Mammoth Black Twig1t Mammoth Black Twig1t Mammoth Black Twig1t

Mann11

Mann?

Mann? McIntosh Red McIntosh Red® McIntosh Red Missouri Pippin® Mother®

alalalalsislalalalcialclclct sisi ce ances cece icici si esi iim iS es Be Be ee ee ee ee ee Be

Pollen

Red June

Williams

Yellow Transparent Chenango

Early Harvest Kinnard

Red Astrachan Red June

Red June

Stayman

Williams

Yellow Transparent Yellow Transparent Ben Davis Jonathan

Baldwin

Ben Davis Northern Spy

Rhode Island Greening}

Doucin Jouathan Newtown

Akin

Ben Davis Early Ripe Jonathan Twenty Ounce Stayman Winesap Montreal Beauty Charlottenthaler Fallwine Pewaukee Maiden’s. Blush Ben Davis Canada Red Opalescent McIntosh Red Tolman Sweet Rome

Stayman Arkansas Black Ben Davis

Ben Davis

Ben Davis Grimes

Rome Spitzenburg Spitzenburg Wagener Winesap Newtown Yellow Newtown Bottle Greening Lady Apple Paragon Hoover’s Red Arkansas Black York Imperial Mann

Red Astrachan Charlottenthaler Hanwell Souring Shiawassee Haas

Pumpkin Russett Ben Davis Lawver Opalescent York Imperial Bonnum

| |

Fruit Set

+ 1: +44) EL

1920.

|No. of Trials

on

242

SeLtF STERILITY AND Cross STERILITY IN THE APPLE. 73

Cross Fertility in the Apple——Continued.

Variety Pollen Fruit Set No. of Trials

Mother® x Stayman +

Newtown?! x White Pippin Sece ate

Newtown?°® x Grimes Golden Paar

Newtown?? x Jonathan +++

Newtown?? x Ben Davis +++

Newtown?? x Spitzenburg SPaRaR

Newtown!+ x Spitzenburg yes

Newtown!?4 x Wagener yes

Newtown? x White Bellflower Para

Nickajack§ x. Stayman — 371 Northern Spy x Baldwin +

Northern Spy x Ben Davis — 2 _ Northern Spy x Golden Russett | —_— 2 Northern Spy x Rhode Island Greening — 2 Oliver’ x Akin +

Ortley11 x Haas yes

Paragon’ x Bloomfield — 60 Paragont12 x Lily of Kent — 46 Paragon® x Lily of Kent +

Paragon?!2 x Stayman =?

Paragon® x Stayman — 157 Paragon® x Stayman — 25 Paragont2 x Winesap — 157 Pewaukee?! x Hoover’s Red yes

Pewaukee?t x Arkansas Black yes

Pewaukee1t x Fallwine yes

Pewaukee11 x Hanwell Souring yes

Ralls® x Northern Spy yes

Red Astrachan x Baldwin — 4 Red Astrachan x Duchess ap

Red Astrachan x Golden Russstt — 2 Red June’ x Early Harvest ++

Red June® x Early Ripe + .

Red June? x Early Ripe ++

Red June’ x Grimes — 35 Red June’ x Williams +

Red June’ x Yellow Transparent ++

Red June? x Yellow Transparent ++

Rhode Island Greening x Baldwin — 2 Rhode Island Greening x Ben Davis — 2 Rhode Island Greening x Golden Russett _— 2 Rome® x Akin — 47 Rome?+ x Ben Davis yes

Romet+ x Newtown yes

Rome& x Northern Spy yes

Romett x Spitzenburg yes

Romes x Stayman — 604 Romet+ x Wagener yes

Shiawasseett x Early Strawberry yes

Shiawasseet x Sweet Bough yes

Shiawassee1t x Tetofsky . yes

Shiawassee B? x Arkansas Black aPaPar

Spitzenburg <A® x Baldwin SPaPap Spitzenburg!+ x Ben Davis yes

Spitzenburg1+ x Jonathan yes

Spitzenburg F°® x Jonathan qParaP Spitzenburg14 x Newtown yes

Spitzenburg E® x Newtown aPAEAP

Spitzenburg A® x Newtown aPaPaP

Spitzenburg B® x Ortley aPar ar

Spitzenburg D® x Red Cheek Pippin SPAR Spitzenburg1+ x Rome yes

Spitzenburg14 x Wagener yes

Stark® x Red Astrachan —_— 84 Staymans’ x Bonnum aF

Staymans x Delicious +

Staymans x Doucin _— 40 Staymans x Early Ripe +

Stayman$’ x Gravenstein _— 300 Staymans’ x Grimes ap

74 Marine AGRICULTURAL EXPERIMENT STATION.

Cross Fertility in the Apple —Concluded.

Variety

Stayman®> Stayman?2 Stayman$’ Stayman? Stayman8s

Stayman5

Stayman$’ Stayman!2 Stayman8’ Stayman?2 Stayman?

Staymans

Steele’s Red11 Steele’s Red12 Steele’s Red11 Summer Permain1t Summer Permain12 Sutton®

Tetofsky11 Tetofsky11 Wagener! Wagener14 Wagener14 Wagener14 Washington?1 Washington!t Washington! Williams’

Williams’

Winesap?1 Winesap13 Winesap?4 Winesap13 Winesap13

Winesap>

Winesap12 Winesap12

Winesap>

Winesap®>

Winesap12

Wolf River®

York Imperial’ Yellow Transparent? Yellow Transparent8 Yellow Transparent® Yellow Transparent® Yellow Transparent® Yellow Transparent® Yellow Transparent8 Yellow Transparent§ Yellow Transparent®

HH HHH KH AK HHH AHH HMM HHH HH KWH MH HH KH MH Od

Pollen

Lily of Kent Lily of Kent Missouri Pippin Nickajack Nickajack Paragon Paragon Paragon Williams Winesap

York Imperial Yellow Transparent Pumpkin Russett Hoover’s Red : Yellow Newtown Salome

Hanwell Souring Northern Spy Mann

Haas

Ben Davis Jonathan

Rome Spitzenburg Oldenburg Hyde’s Keeper Charlottenthaler Early Ripe Stayman

Yellow Transparent Arkansas Black Ben Davis

Ben Davis Grimes Jonathan

Lily of Kent Lily of Kent Paragon Paragon Stayman Stayman

Yellow Transparent Missouri Pippin Early Ripe Early Ripe Nickajack

Oliver

Red Astrachan Red June

Stark

Stayman Williams

Fruit Set

Sods eaacesecaltitiiit+its

1920.

No. of Trials

14

35 212

Table 4 shows that of the 243 tests for cross sterility be- tween two varieties 57 are recorded as not producing fruit, 186 as producing fruit of which go produced fruit but did not re- cord the number of crosses made to accomplish its production. These figures show that over 34 of the varieties crossed proved

compatible with each other.

It will be remembered that nearly

24 of those which were self fertilized showed no fruit produc- tion. These facts argue strongly for the necessity of arranging

SELF STERILITY AND Cross STERILITY IN THE APPLE. 75

for cross pollination in the commercial production of apples. If the relative set of the fruit is considered it is even more clear- ly demonstrated that cross pollination is necessary in commer- cial orcharding for of the 42 self fertilized which did set fruit as shown in table 2, less than 16 set fruit in anything but neg- ligible amounts.

It is of some interest to examine the crosses which did not set fruit a little further to determine if possible the reason why they did not. Out of the 57 which did not prove compatible about half (26) had enough trial crosses made to make it seem likely that these crosses were nearly if not entirely, incompat- ible. These crosses were Delicious x Grimes, Duchess x Bald- win, Early Ripe x Chenango, Early Ripe x Kinnard, Ingram x Rome, Ingram x Stayman, Lily of Kent x Paragon, Nickajack x Stayman, Paragon x Bloomfield, Paragon x Lily of Kent, Paragon x Stayman, Paragon x Winesap, Red June x Early Ripe, Rome x Akin, Rome x Stayman, Stark x Red Astrachan, Stayman x Doucin, Stayman x Gravenstein, Stayman x Lily of Kent, Stayman x Missouri Pippin, Stayman x Paragon, Stay- man x Winesap, Winesap x Lily of Kent, Winesap x Paragon, Yellow Transparent x Stark, and Yellow Transparent x Stay- man. It will be noted that the varieties Stayman, Winesap and Paragon form the largest part of these sterile crosses. Stay- man is known to be a seedling from the Winesap.* The Para- gon is thought to have originated from the Winesapy+ crossed by Limbertwig. If these facts represent the true state of af- fairs it is entirely likely that the seedlings would also have the incompatibility of the parents from which they sprang pro- vided, of course, that sterility in the apple is inherited in a sim- ilar manner to other known inheritance.

It is of interest to note also that the variety Lily of Kent enters into a number of these crosses. Lily of Kent x Paragon and Paragon x Lily of Kent are reciprocally sterile. Lily of Kent pollen is also sterile with Stayman and Winesap. So, likewise, is the cross between Yellow Transparent x Stayman and Stayman x Yellow Transparent reciprocally sterile. On the other hand the crosses of Nickajack x Stayman and Red June x Early Ripe are sterile but the reciprocal crosses are

*Beach, S. A., et al., 1905 The Apple of New York. vol. I, p. 318. tBeach, S. A., et al, 1905 The Apple of New York. vol. I, p. 247.

76 MAINE AGRICULTURAL EXPERIMENT STATION. 1920.

fairly fertile and produce fruit. Crosses, Stayman x Doucin and Stayman x Gravenstein are sterile but the cross Gravenstein x Doucin is fertile. These facts make it clear that because a cross between two varieties (a x b) is sterile it is no guarantee that the reciprocal cross (b x a) will be sterile. Further if the cross of two given varieties (a x b) is sterile and the cross of two varieties including one of the given varieties (a x c) is sterile it is apparently equally possible for the two different va- rieties entering into the cross (b x c) to be compatible or in- compatible. The varieties which are particularly fertile when crossed are of especial interest to the man who desires to plant a com- mercial orchard or to increase the bearing ability of one already in existence by top working certain of the trees. Those crosses which are marked with the three pluses (+-+-+) in table 4 should prove heavy bearers when planted together. Such or- chards should be planted with the female parent, indicated in the first column, as the predominating tree in the block. Among the leading varieties in Maine which should form desirable combinations for commercial work are Baldwin with the Golden Russett for the pollen parent; Ben Davis with McIn- tosh Red, Northern Spy, Opalescent or Wealthy for pollen par- ent; Golden Russett with Ben Davis for the pollinator. Esopus can be planted with Ben Davis and Jonathan. Newton crosses well with any of the common pollen varieties Grimes Golden, Jonathan, Ben Davis or Spitzenburg. The relative compatibil- ity of the other varieties may be seen by consulting the lists. The work of Alderman* makes it clear that the differences in the yield of the fruit in self and in cross pollinated orchards occupies about the same relations as are shown in the hand self pollinations of table 2 and the hand cross pollinations of table 4. In this experiment a Rome Beauty* orchard that had been bearing only moderate crops was cross pollinated by bringing in branches of other varieties and allowing the bees to work over these other varieties at the same time that they worked over the Rome Beauty. A suitable control was made with an-

*Alderman, W. H., 1917. Experimental Work on Self-sterility of the Apple. In Proc. Amer. Soc. for Hort. Sci. p. 94-101.

*The Rome Beauty as will be seen in table 2 is nearly if not quite self-sterile. ;

Setr STERILITY AND Cross STERILITY IN THE APPLE. 77

other block of Rome Beauty trees some distance away. The

cross fertilized Rome Beauty trees yielded 174% bushels; the

check Rome Beauty trees for which no arrangement for cross

fertilization was made, yielded 83% bushels or the cross fertil-

ized trees had nearly twice the yield of the other check block.

The demonstration was made complete by a repetition of the

experiment in a succeeding year.

THe GROWTH VIGOR AND RESULTING SIZE OF APPLES FROM.

SELFED OR CROSSED VARIETIES.

Certain objections may be made to the introduction of cross.

fertilization on the ground that where such cross. fertilization takes place a scrub is produced which is worse than either par-

ent. If such is the case it would be the height of folly to cross

pollinate even though there was an increased yield, for apples

are largely sold on the basis of their color, shape and size, and.

if these items are not properly developed the increased yield

would not make up for the reduced selling price. The data in-

table 5 present the material to analyze this problem.

TAB TEES:

Size and Number of Seed from Selfed and Crossed Fertilized

_ Apple Blossoms.

Baldwin x Baldwin 5 6.36 128 ¢—22 p. Baldwin x Duchess 1 7.00 | 4.0 g —3.0 p. Baldwin x Golden Russett 48 6.36 | 3.8 g—2.8 p. Ben Davis x Canada Red 8 6.65 | 5.5 g—1.4 p. Ben Davis x Crab 12 6.25 44 ¢—14p. Ben Davis x Golden Russett 5 | 5.66 | 6.2 g—0.8 p. Ben Davis x Gravenstein 3 | 6.53 1.7 g—0.3 p. Ben Davis x McIntosh Red 74 6.28 | 6.2 g—1.1 p. Ben Davis x Northern Spy 14 | 5.22 | 5.9 ¢g—0.4D. Ben Davis x Opalescent 35 6.38 7.0 g—0.8 p. Ben Davis x St. Lawrence 6 6.75 6.1 ¢g—0.2 p. Ben Davis x Wealthy 16 4.94 ZEB} 8} ie Early Harvest x Duchess 2 Soca | 8.0 g—0.0 p. Golden Russett x Ben Davis 5 6.22 | 7.4 ¢—1.2 D. Golden Russett x Northern Spy 3 5.40 | 7.7 ¢—0.0 p. Hurlbert Sweet x Ben Davis | 3 7.17 | 4.3 ¢g—3.7 D. Hurlbert Sweet x Canada Red 1 6.70 | 3.0 g—3.0 D. McIntosh Red x Opalescent | 2 6.15 | 6.5 ¢—0.5 p. Wealthy x Wealthy | 1 aon | 4.0 g—0.0 D.

lee

7 ES: lal a

78 MAINE AGRICULTURAL EXPERIMENT STATION. 1920.

Data on the size and number of seeds of the apples result- ing from a cross are presented in summary form from appendix table I.

From the data contained in table 5 it is clear that the Bald- win apples resulting from cross pollination were of as good average size as were the apples which resulted from self fertili- zation. Since the set of fruit from the cross fertilization was larger than from the self fertilization it follows that the profit to the grower was much greater for the blossoms where cross pollination took place than where self pollination was resorted to.

The apples resulting from cross pollination of the Ben Davis were likewise all of good size from the market stand- point, as were also the apples from the other crosses. They carried more good seeds than did the self fertilized apples. From these facts we may conclude that the size of the fruit is favor- ably affected rather than otherwise by cross pollination.

The amount of this cross pollination affect appears to dif- fer with different varieties. Alderman, W. H.* found that for the Rome Beauty above mentioned the cross pollination by other varieties increased the size (weight) 27.8 per cent over that of the apples resulting from self fertilization. For York Imperial the increased size for cross pollination was 42.7 per cent over the size of the selfed apple. For Wagener the effect of cross fertilization over self fertilization was in the direction of reduced size the reduction being 17.3 per cent. The results of these ex- periments would seem to show in general a beneficial effect of cross fertilization on size. Some work of Wicks, W. H.7 using reciprocal crosses of the Ben Davis, Grimes, Jonathan and Winesap varieties to determine the effect of crossing versus selfing on the resulting color, size and quality of the fruit quite clearly shows that for these items the characters of the Mother parent varieties are found in the resulting fruit irrespective of what pollen parent is used.

*Alderman, W. H. 1917. Experimental Work on Self-Sterility of the Apple. In Proc. Amer. Soc. for Hort. Sci. p. 94-101.

TWicks, W. H., 1918. The Effect of Cross Pollination on Size, Color, Shape, and Quality of the Apple. In Bul. 143. Arkansas Agr. Expt. Station.

SeLF STERILITY AND Cross STERILITY IN THE APPLE. 79

It is true that certain differences may be noted dependent upon the pollen supplied for a given cross. These differences are not in immediate relation to the variety of pollen supplied, but depend upon complex factors which will be analyzed in sub- sequent publications. Furthermore the effect of the crosses may be toward increased color in one cross and decreased color in another, etc. So far as the effect on the fruit is concerned it is absolutely safe and advisable to plant two varieties of dif- ferent color, shape, etc. together. A red apple will be just as red if pollinated with pollen from a green variety as if pollinated with a red pollen variety. Of course the seeds resulting from such crosses will be different in the two cases, but the flesh or marketable portion will remain unchanged.

This conclusion would be expected from other independent evidence taken from histological studies of the development of the apple. The apple is like an enlarged branch of the mother tree. It does not receive anything of a genetic nature from the resulting union of the pollen and the ovule. It only acts like a sack to protect the seed. It is all maternal in origin and would therefore be expected to assume the maternal characters, size, Shape, quality and color, of the mother tree.

If we look at the problem in the light of the preceding data ‘on the self sterility and the cross sterility of the different varie- ‘ties it is found that the number of fruit set from self fertiliza- ion is so limited as to make it entirely likely that the large pro- portion of the apples in commercial orcharding are the result of cross fertilization. Thus in table 2, one of the best com- mercial varieties, the Baldwin, matured on self fertilization 20 fruit out of 409 trials, a percentage of about 5. On cross fertili-_ zation this variety produced good fruit in something over 50 per cent of the crosses which were made. The Ben Davis variety matured no fruit in Maine on self fertilization yet this variety is capable of bearing a crop of a color and size consistant with the best of the variety even though the majority of fruit must have been formed by cross fertilization with a foreign pollen. In view of what the investigations on the causes of self sterility have shown in relation to the growth of the pollen tube it would seem more probable that in the commercial orchard the percent- age of fruit set from self fertilization would be considerably ‘below the percentage obtained in experimental work. Thus

80 MAINE AGRICULTURAL EXPERIMENT STATION. 1920.

given an even start the growth of the pollen tube in the style of the compatible pollen is so rapid as compared with the growth of the pollen tube of the incompatible pollen that in the major- ity of cases the compatible pollen would beat out the incom- patible pollen in the fertilization of the ovule. Such a competi- ° tive race is, of course, eliminated in experimental work where the incompatible pollen and that only is allowed to grow in the style. Should it be assumed, however, that the number of fruit matured for the other stations is more representative of the percentages matured for the Maine Ben Davis orchards even this percentage (it is only about 1.5) will not account for the crop of fruit obtained in some of the favorable apple years, when this fruit is all of excellent size and color. These facts all strengthen the conclusions as expressed above and as demon- strated by controlled experiment in Arkansas that the color of the fruit, the size and other characteristics of the variety are as pronounced in the apple resulting from cross fertilization as they are from the apple resulting from self fertilization.

It may therefore be safely concluded that the data on cross fertilization in the apple show that an increased yield results and the size, color and quality of the apples are equal to those from self pollination. To be commercially desirable an orchard should, therefore, be a mixture of the varieties which have com- patible pollen.

This conclusion may seem contrary to what is considered good commercial practice which has in the past favored large blocks of a single variety of apple. As shown above by results only recently determined, the apple tree must be crossed fer- tilized to produce good, regular crops of commercially desirable fruit. By this it is not meant that an orgy of promiscuous re- grafting or planting of many varieties in one block is advocated. It means simply that two varieties which are reciprocally com- patible should be planted together. The trees for pollination may be reduced to a minimum of only 5 per cent or one tree in 20. In planting every fourth tree in each fourth row is the pollenizer to accomplish this result. Promiscuous grafting is likewise bad commercially since it makes harvesting especially difficult. If it is desired to grow the varieties in equal propor- tions alternate blocks of not more than 4 or 5 rows may be

SeLr STERILITY AND Cross STERILITY IN THE APPLE. 81

planted. In any case not more than 4 or 5 rows should separate the pollenizer trees from those to be pollinated.

For orchards already planted, regrafting a desirable pol- lenizer in the above mentioned proposition may be practiced. While waiting for these pollenizers to grow to bearing age a practical relief may be had by cutting large branches of other good pollinating varieties and placing them in water pails hung from the tree limbs.

Experiment has shown that little pollen fertilization is brought about by wind. Insects, wild and cultivated are the

best agents to transport pollen from one variety to another. It is therefore commercially profitable to keep bees in the orchard

for this purpose even though no honey is produced.

CAUSES OF SELF STERILITY AND Cross STERILITY.

Sterility within the different species of plants appears to be due to several causal agents. These agents may be external or they may be internal. The external agents include such things as disease affecting the vitality of the tree or its blossoms such as scab, fire-blight, insect infections, spray injury before, during or after flowering. Low temperature and cold continued rains at flowering time may be other factors determining the amount of fruit set and consequently its yield. These factors are more or less under the control of the apple grower and should receive careful attention. They need not be discussed here for the remedial agents are well known.

The internal causes for sterility include degenerate pollen; pollen which is not able to cooperate properly with the style to facilitate the growth of the pollen tube at a sufficient rate of

‘growth to reach the ovule and cause fertilization; and lack of

proper development of ovule.

Within the apple the phenomena of self sterility is appar- ently quite universal. The crosses of the varieties which are self sterile with pollen which is crossed fertile with them show that the ovules are capable of fertilization and are therefore not responsible for the sterility resulting from the self fertilization. Similarly the argument could be made that since the pollen from a self sterile variety is capable of fertilizing other varieties the

eet

a a

82 Maine AcricuLtuRAL ExPERIMENT Station. 1920.

pollen as such is not responsible in self sterile varieties of apples for the fruits not setting.

Investigation shows that the problem is one of the interre- lation between the pistil and the pollen andthe pollen tube. It has been shown that in the self sterile varieties self fertilized the pollen tube grows much more slowly than does the pollen tube of other varieties of pollen when used on the same pistils.* Thus in the self fertilized flower the rate of growth of the pollen tube is so slow that it cannot traverse the length of the style and fertilize the ovule before the ovule withers and dies. With the cross pollinated flowers the pollen tube grows much more rapidly and easily reaches the ovule in time for fertilization to take place. The physical basis of one form of this sterility is consequently due to some factors which inhibit the growth of the pollen tube in the style of the same variety. What this dif- ference is, is a matter now under further investigation.

LITERATURE LIST.

1. Waugh, F. A. 1900. Report of the Horticulturist. In Thirteenth Annual Report of Vermont Agricultural Experiment Station. 1899-1900. p. 364. 2. Powell, G. Harold. 1901. Report of the Horticulturist. In Thirteenth Annual Report of the Delaware Agricultural Experiment Station, 1901, p. 114. |

1900. Report of the Horticulturist. In Twelfth Annual Report of

the Delaware Agricultural Experiment Station, 1900. p. 134-139.

4. Lewis, C. I. and Vincent, C. C. 1909. Pollination of the Apple. In Bul. 104 Oregon Agricultural Experiment Station, pp. 19-20.

5. See reference 3.

6. Hedrick, W. P. and Wellington, Richard. 1912. An Experiment in Breeding Apples. In Thirty-First Annual Report of the New York Experiment Station, p. 457.

7. Ballard, W. R. 1916. Methods and Problems in Pear and Apple Breeding. In Bul. 196. Maryland Agricultural Experiment Station. p. 88.

8. See reference 7.

9. See reference 4.

*Knight, L. I., 1917. Physiological Aspects of Self-Sterility of the Apple. In Proc. Amer. Soc. Hort. Sci. p. 101-105.

10. Mal 12:

13.

14.

SELF STERILITY AND Cross STERILITY IN THE APPLE. 83

See reference 4.

See reference 11.

Closes. G2. P-

1902. Report of the Horticulturist. In Fourteenth Annual Report of the Delaware Agricultural Experiment Station, p. 102.

Wicks, W. H.

1918. The effect of cross pollination on size, color, shape, and qual- ity of the apple. In Bul. 143. Arkansas Agricultural Expt. Sta. Vincent, C. C.

1915. Report of the Department of Horticulture. In Bul. 84, Idaho Agricultural Experiment Station, p. 24.

84 Maine AGRICULTURAL EXPERIMENT STATION. 1920.

APPENDIX TABLE 1.

Apples Resulting from Selfing and Crossing of Varieties.

Parents Selection Diameter No. Seeds Number | Mother Parent Pollen 146 Baldwin x Baldwin 5.6 cm. 2 —— 2) PP 230 Baldwin x Baldwin 5.7 em: 6 g—2 P 231 Baldwin x Baldwin 6.6 cm. —6 P 248 Baldwin =x Baldwin = 6S een 3 — 249 Baldwin x Baldwin TA. em: 3 g¢g—1-P Average 6.36 ¢m 2.8 ¢ —2.2 P 147 Golden Russett x Ben Davis 6.6 em.| 10 ¢g 148 Golden Russett x Ben Davis 6.7 cm. 9 ¢g 149 Golden Russett x Ben Davis 6.6 em. 7 g¢—1 P 152 Golden Russett x Ben Davis 5.8 cm. 8 g— 150 Golden Russett | =x Ben Davis 5.4 em. 3 g¢—5 P Average 6.22 em. 74 ¢—12P 16 Hurlbert Sweet x Ben Davis led meres 5 g—4 P 15 Hurlbert Sweet x Ben Davis . 42) ene 4 g¢—4 Pp 13 Hurlbert Sweet x Ben Davis 7.1 cm. eee) ee: Average : 7.17 em. 43 ¢—3.7 P 7 Ben Davis x . Canada Red 62- em.) 26. -¢/—1T- P g Ben Davis x Canada Red 5.8 cm. 42/2 P 9 | Ben Davis x Canada Red 6.6-em. | 62 —1- Pe 10 |} Ben Davis x Canada Red | 65 cm: 8 g—1 P il Ben Davis =x Canada Red | 7.0 cm. 4_-g¢—4 P 12 Ben Davis x Canada Red P23, -CM e622 —— alae. 13 Ben Davis x Canada Red 7.0 cm. 6 £¢— 14 | Ben Davis x Canada Red 6.8 cm. 4 ¢—1 P Average 6.65em. 552g —14P 3 Hurlbert Sweet ~ x Canada Red 6.7 em.| 3 g—3 P 20 Ben Davis x Crab 5.7. cm. eee 22 Ben Davis x Crab 6.3 cm. 5 g—I1 23 Ben Davis x Crab 6.2 em. 4 ¢g S 26 Ben Davis x Crab 6.7 cm. 6 g—1 P 27 Ben Davis x Crab 6.4 em. Ms SP 29 Ben Davis x Crab 6.6 em. 6 ¢g 30 Ben Davis x Crab 7.3 cm. AW 5g 19 Ben Davis x Crab 6.0 cm. 3 g¢g—4 P 21 Ben Davis x Crab 5.4-em.| 1 ¢g—4 P 24 Ben Davis x Crab 5.9 cm. 2 3 | P 25 Ben Davis x Crab 5.8 cm. 3 g¢g—2 P 23 Ben Davis x Crab 6.7 cm. Das — iS P. Average 6.25 em. 44 ¢—14P 153 | Baldwin x Duchess 70 cm.|-45¢=—38) Pe 144 Early Harvest x Duchess 2? 9 ¢g 145 Early Harvest =x Duchess rg Moc Average 2 8 g 163 | Baldwin x Golden Russett aye tre 6 ¢g 164 | Baldwin x Golden Russett 6.0 cm. Dag 165 Baldwin x Golden Russett 6.5 cm. Dee) IP. 166 Baldwin x Golden Russett 5.5 em. 3 g—4 P 167 Baldwin x Golden Russett 6.3 em. 2g — FP 168 Baldwin x Golden Russett 6.8 cm. 2 ¢—6 P 169 Baldwin x» Golden Russett 5.5 cm. ee = ee: 170 : Baldwin x Golden Russett 6.1 em. fo —— ee 171 | Baldwin x Golden Russett 6.3 cm. 6 g¢g—3 P

i, ;

Setr STERILITY AND Cross STERILITY IN THE APPLE.

Apples Resulting from Selfing and Crossing of Varieties.

—Continued. Parents Selection | Number Mother Parent Pollen 172 Baldwin x Golden Russett 173 Baldwin x Golden Russett 174 Baldwin x Golden Russett 175 Baldwin x Golden Russett 176 Baldwin x Golden Russett 177 - Baldwin x Golden Russett 178 Baldwin x Goiden Russett 179 Baldwin: x Golden Russett 180 Baldwin x Golden Russett 181 ‘Baldwin x Golden Russett 182 Baldwin x Golden Russett 183 Baldwin x Golden Russett 185 Baldwin x Golden Russett 186 Baldwin x Golden Russett 187 Baldwin x Golden Russett 188 Baidwin x Golden Russett 190 Baldwin x Golden Russett 191 Baldwin x Golden Russett 192 Laldwin x Golden Russett 193 Baldwin x Golden Russett 194 Baldwin x Golden Russett 195 Baldwin x Golden Russett 196 Baldwin x Golden Russett 197 Baldwin x Golden Russett 198 Baldwin x Golden Russett 200 Baldwin x Golden Russett 201 Baldwin x Golden Russett 202 Baldwin x Golden Russett 203 Baldwin x Golden Russett 204 Baldwin x Golden Russett 205 Baldwin x Golden Russett 207 Baldwin x Golden Russett 208 Baldwin x Golden Russett 209 Baldwin x Golden Russett 210 Baldwin x Golden Russett 184 Baldwin x Golden Russett 189 Baldwin x Golden Russett 199 Baldwin x Golden Russett 206 Baldwin x Golden Russett Average 227 Ben Davis x Golden Russett 228 Ben Davis x Golden Russett 229 Ben Davis x Golden Russett 242, Ben Davis | x Golden Russett 243 * Ben Davis | x Golden Russett Average 4 Ben Davis x Gravenstein 5 Ben Davis | x Gravenstein 6 Ben Davis | x Gravenstein | Average 31 Ben Davis | x MeIntosh 32, Ben Davis | x MelIntosh 33 Ben Davis | x McIntosh 34 Ben Davis | x MeIntosh 35 Ben Davis | x McIntosh 36 Ben Davis | x McIntosh 37 Ben Davis | x MelIntosh 38 Ben Davis | x MeIntosh

D | DRABAAARAAARAAAAATAAHAAHIN DHA AHH HAD UA HAH DOAN DP to] WINDS OONM UDR WINTWNUADOAVWRUNTARNNORRODONONWNIO

| woe fo) for Sil a

a] DN

ow

WOaHOmaONLS

DARRAMAD B| AAD 1W| ganTds

Diameter

QwWwwonwidtnwo HH] NNR FB] atwP Oo VIONDONPNNHWUWANNNWNHNWRNW DOBPAHWHWTIAHP WOT

85

No. Seeds g—1 P g—2 P g—3 P g~—4 P g¢—3 P g¢—l1 P a g—3 P g—3 P g—4 P g—3 P g—2 P eS = g—8 P

—4 P g¢—3 P g¢—l1 P zg—4 P g—1 P g—6 P g¢—4 P g¢—s P g¢—5 P g—5 P g—2 P g—1 P g—5 P == g—4 P g¢—6 P g—3 P g—6 P g—4 P g—3 P eS = => f= 8ig — 2.8 P g—2 P g

g

g—2 P 2¢—08P g—1 P g

£ 7¢—0.3P 4

g¢—l1 P g—1 P g

g—1 P 2) — 2) .g—4 P g—4 P

86 Marne AGRICULTURAL EXPERIMENT STATION. 1920.

Apples Resulting from Selfing and Crossing of Varieties.

—Continued. Parents Selection ‘Diameter, No. Seeds Number Mother Parent Pollen $f cana PE

39 Ben Davis x McIntosh 6.6 em. 5a ge

47 Ben Davis x McIntosh |65 em. 7 g—2 P 48 Ben Davis x McIntosh 64 em.| 7 g—2 P 49 Ben Davis x McIntosh 69 em.|. 7 £ : 50 Ben Davis x McIntosh 6.5 em. je — so & 51 Ben Davis x McIntosh 6.7 em.|} 8 g—1 P 52 Ben Davis x McIntosh | 6.8 em. CS

53 Ben Davis x McIntosh {58 em.| 5 g—3 P 54 Ben Davis x McIntosh 165 em.| 8 g—2 P 5D Ben Davis x McIntosh /60 em.) 7 g—

56 Ben Davis x McIntosh VEG rcmai 6 -f— 1 57 Ben Davis x McIntosh |'65 em.| 8 g—2 P 58 Ben Davis x McIntosh Wiper 11S [ie 1 fe a or 59 Ben Davis x McIntosh |61.em.| 4 g—1 P 60 Ben Davis x McIntosh |62 em.| 2 g—4 P 61 Ben Davis x McIntosh /65 em.| 7 g—1 P 62 Ben Davis x McIntosh 6335 cm | ee Saee— OP 63 Ben Davis x McIntosh | 6.7 cm. | 9 g— 64 Ben Davis x McIntosh }69 em.|, 7 g—1 P 65 Ben Davis x McIntosh |/66 em.) 6 g—1 P 66 Ben Davis x McIntosh |66 em. 6 g— 67 Ben Davis x McIntosh 63 cml vee — ae oP 68 Ben Davis x McIntosh 7.0 cm. 6 g—1 P 69 Ben Davis x McIntosh 61 em.; 8 g 70 Ben Davis x McIntosh 58 em. 8 g—1 P 71 Ben Davis x McIntosh 5.6 em. Tia bP 72 Ben Davis x McIntosh 49 em.| 1 g—5 P 73 Ben Davis x McIntosh Ga fny|| GW ae

74 Ben Davis x McIntosh 63 em.| 6 g—1 P 75 Ben Davis x McIntosh 61 em.) 6 g—

76 Ben Davis x McIntosh 58 em 8 g—

77 Ben Davis x McIntosh 59 em. 6 g=2

78 Ben Davis x McIntosh 6.1 em.; 5 g—1 P 79 Ben Davis x McIntosh 478 em.| 9 £—

80 Ben Davis x McIntosh 64 emesis 2 SP 81 Ben Davis x McIntosh 66 em.) 8 g— 82 Ben Davis x McIntosh 70 em. 7 £— 83 Ben Davis x McIntosh 6.7 em. 9 g— 84 Ben Davis x McIntosh 65 em.| 7 g—1 P 85 Ben Davis x MeIntosh 64 cm.| 7 g—1 P 86 Ben Davis x McIntosh 6.4 em.|' 6 ¢g 87 Ben Davis x McIntosh 68 em.| 5 g—1 P 88 Ben Davis x McIntosh 6.5 em.| 8 g

89 Ben Davis x McIntosh 58 em.| 3 -—3 P 90 Ben Davis x McIntosh 53 em.| 8 ¢g

91 Ben Davis x McIntosh 5.9 em.| 8 g

92 Ben Davis x McIntosh 65 em.| 5 g—1

93 Ben Davis x McIntosh 65 em.’ 6 g—1 P 94 Ben Davis x McIntosh 6.3 cm. Lee

95 Ben Davis x McIntosh 5.6 em.| 7 g

96 Ben Davis x McIntosh 7.0 em. 8 g

> 3 led go ny a & Qa bo oQ | eH at ty

SELF STERILITY AND CROSS STERILITY IN THE APPLE.

Apples Resulting from Selfing and Crossing of Varieties.

—Continued. Parents Selection Diameter No. Seeds Number Mother Parent Pollen 151 Ben Davis x Northern Spy 5.4 em.| 5 g— 154 Ben Davis x Northern Spy 5.9 em. 6 ¢g 156 Ben Davis x Northern Spy 4.7 cm. OW (e 157 Ben Davis x Northern Spy bn Gil) 3 fS 158 Ben Davis x Northern Spy 5:2) ‘em: B) FR 159 Ben Davis x Northern Spy 5.2 em. 9 g 244 Ben Davis x Northern Spy 5.4 em. 38 g—1 238 Ben Davis x Northern Spy 5.2) em: Wf fs 237 Ben Davis x Northern Spy HY) Gal | Ve fs 234 Ben Davis x Northern Spy Bua Gam, |) 9B) fA 233 Ben Davis x Northern Spy 45 em.| 5 g—83 247 Ben Davis x Northern Spy 3.9 em. 4 ¢g 250 Ben Davis x Northern Spy Bh Om || Fs 251 Ben Davis x Northern Spy Gal (an || by FS Average 5.22 em 5.9 g — 0.4 160 Golden Russett x Northern Spy i Gans |) Ff {3 161 Golden Russett x Northern Spy 5.6 em. ete 162 Golden Russett x Northern Spy ? 9 g Average Bye! nk) WHO ES 103 Ben Davis x Opalescent 6.7 cm. 6 g—3 104 Ben Davis x Opalescent 6.1 em. 3 g—6 105 Ben Davis x Opalescent Te Gin, 3 4 106 Ben Davis x Opalescent 6.1 em. 7 g—1 107 Ben Davis x Opalescent 6.0 em. ue ES 108 Ben Davis x Opalescent Ber (onal, |) ah tS 109 Ben Davis x Opalescent 62 em.| 6 g—1 - 110 Ben Davis x Opalescent 5.9 em. 6 g—1 111 Ben Davis x Opalescent 7.1 cm. Sak 112 Ben Davis x Opalescent 5.8 em. 5 g—2 113 Ben Davis x Opalescent 5.6 em.| 12 ¢g 114 Ben Davis x Opalescent 5alliem: 9 g— 71 115 Ben Davis x Opalescent 66 em.| 8 g 116 Ben Davis x Opalescent Gl @im,|) 4 & 117 Ben Davis x Opalescent 6.5 em. @ 118 Ben Davis x Opalescent 6.8 ecm. YS 119 Ben Davis