de;partme;nt of botanical research. 6;^ 



condition of the trichomes of both hybrids. In Solanum four types of trich- 

 omes are to be found, of which one is common to both pure hues and is to be 

 seen in the cross also ; two are peculiar to Solanum villosuni and do not occur 

 in the hybrid in either the first, the second, or the third generation, and the 

 fourth, which is found in Solanum guinense and not in the other pure line, 

 appears undiminished in size in the cross. The trichomes of Juglans, of which 

 there are five types, are with one exception multicellular and each type has its 

 peculiar life-history, which is adhered to with great, perhaps with perfect, 

 consistency. This fact, should it be found to be of general application, would 

 suggest the use of multicellular trichomes whenever possible or desirable as 

 an aid in determining the relationships of closely related species and as sup- 

 plemental to embryogeny proper. A comparative study of the different types 

 of trichomes of Juglans shows certain developmental sequences which are ex- 

 perienced in common by them all, so that the mutual relationships of the dif- 

 ferent forms can be investigated and possibly their structural origin can be 

 traced. In all pure species and hybrids the trichomes were found to vary 

 considerably in size, but this variation was seen to be closely associated with 

 the position occupied by the trichome on the plant member. From the high 

 degree of consistency which the variation exhibited as related to position, as 

 well as because of the wide range of the phenomenon, it was concluded that 

 the minor fluctuations were in some manner to be associated with an unequal 

 food supply, or nutritive conditions, rather than with direct transmission of 

 such qualities from the pure lines, or with the incident of the cross. Through- 

 out the course of the investigation, watch has been kept for structural evi- 

 dences of unit-characters, or of anatomical attributes of unit-characters. 

 Although no conclusive results were derived, certain suggestions were ob- 

 tained which ought to be of service in any investigation on this general topic. 

 For instance, in Oenothera it was evident that the trichomal system could not 

 be taken in its entirety and be properly considered a single character, since 

 each of the three types had a different and apparently an independent distri- 

 bution. Again, in Juglans it is possible that the origin of a unit-character can 

 be traced in the origin and differentiation of the different related types of 

 trichomes. Such observations point to the need, where the inheritance of 

 qualities is being carefully studied, of closer anatomical investigations than 

 are usually made. 



Root-hahits. — In a research on root-habits, which is being made by Dr. 

 Cannon, it is proposed to include the leading types of plants which occur nat- 

 urally in the vicinity of Tucson, and so far as possible to extend the study of 

 plants of other arid regions for purposes of comparison, as well as for the 

 purpose of extending our knowledge of the roots of plants. The studies will 

 include the most important features of the external and the internal morphol- 



