Department of Experimental Plant-Breeding. liii 



as yet imperfectly understood, are of great importance. Consid- 

 erable work of this nature has been started and includes the study 

 of Mendelian segregation and recombination in connection with 

 such characters as form, size, flavor, color, etc. Experiments are 

 being conducted by Mr. A. W. Gilbert on tomatoes, phloxes, sal- 

 piglossis and morning glories, with special reference to the laws 

 of inheritance governing the transmission of color and shape. In 

 connection with these experiments, about 6,000 tomato plants and 

 several thousand phlox plants have been grown during the last 

 summer. 



Other work of similar nature is being conducted by the depart- 

 ment with peppers, carnations and verbenias. In this work, while 

 the aim is purely scientific, as a secondary result forms of some 

 value for propagation may be found, and in such cases the investi- 

 gators will expect to make use of them. 



Studies of mutations and their use in breeding. — The studies 

 of mutations which have been conducted by various investigators 

 during the last few years, if reliable, would seem to necessitate 

 changes in our general method of breeding, and it is important at 

 the present time to determine the extent to which variations of 

 this nature appear and their true value in connection with practical 

 breeding. As rapidly as possible studies of native vegetation are 

 being made to determine how commonly such variations are pro- 

 duced in nature. In this connection, investigations are being made 

 on a limited scale with dandelions, plantago, daisies, buttercups, 

 silene, etc. The most important investigation on this subject, how- 

 ever, is that conducted in connection with the timothy breeding 

 experiments. Careful studies are being made of the numerous 

 striking variations which have been found to occur in timothy and 

 which are probably to be interpreted as mutations that have been 

 produced under the influence of cultivation. This study, while 

 closely connected with the practical breeding experiments of 

 timothy as above discussed, is nevertheless of great interest from 

 the scientific standpoint in the study of mutations, and is there- 

 fore being pursued more carefully than would be required were 

 we dealing only with the production of new varieties of value. 



Studies on the cumulative action of selection. — The claims of 

 DeVries with particular reference to the breeding of self-fertilized 

 plants, as stated in his Plant Breeding, serve to throw doubt on 

 our old methods of selection. Ever since tlie appearance of 

 Darwin's Origin of Species, and even before this time, it was 



