xlvi Department of Experimental Plant-Breeuing. 



II. investigation. 



The support of this department is derived ahiiost wholly from 

 Federal funds and the work of the department is limited, therefore, 

 almost entirely to investigation of problems connected with plant- 

 breeding. 



As the department was organized in April, 1907, the experiments 

 for the most part have been under way for only two seasons, so 

 that the greater part of the investigations are still incomplete. 



Timothy breeding experiments. — The timothy breeding experi- 

 ments are conducted by the writer with the assistance of Dr. C. F. 

 Clark. 



Timothy is one of the most important crops grown in New York 

 and is deserving of carefr.l study to determine whether the crop 

 may be improved in yield and in quality. The necessity of breed- 

 ing improved varieties of timothy was recognized by the Experi- 

 ment Station several years ago, and extended plantings of timothy 

 seedlings were made to furnish foundations for starting selections. 

 The early details regarding these experiments were published in 

 the last annual report of this department. L'p to the time the 

 writer took charge of these experiments, plants to the number of 

 17,220 had been planted under controlled conditions, one plant in 

 a place, the plants several feet apart, like corn. This large number 

 of individuals furnished an excellent opportunity to distinguish dif- 

 ferences and make selections of desirable types. Over 200 different 

 types or forms were selected in 1907 and these were planted in 

 short rows by means of slips or suckers, to test their characters 

 more fully and to obtain a more reliable judgment as to their com- 

 parative value. These plats have this last summer reached a stage 

 of development in which the different types show their character- 

 istics very plainly. The results show conclusively that we have in 

 these selections a very wide range of characters. Some of the 

 types, indeed, are so distinct that they could almost be considered 

 distinct species. In the character of yield, for instance, three low- 

 yielding plats gave an average yield per plant of .06, .13 and .28 

 of a pound respectively. Three similar high-yielding plats, grown 

 under the same conditions, gave an average yield per plant of 1.14, 

 I. GO and .97 pounds respectively. In these six plats, the three high- 

 yielding plats gave nearly seven times the yield of the three low- 

 yielding plats. It has been found further, by observation, that the 

 high-yielding, more vigorous strains start much earlier after cut- 

 ting, so that they would furnish more fall pasturage and their com- 



