DEPARTMENT OF BOTANICAL RESEARCH. 99 



though distinct and clear-cut, but in most cases the differences are so 

 pronounced and conspicuous as to be recognizable at considerable dis- 

 tances in the field. The differences exist in practically every character 

 of the plant, such as size, color, arrangement of the flowers, time of 

 flowering; shape, size, color, arrangement of leaves, general habit of 

 the plant, as erect, diffuse, open, compact; ranging from tree-hke erect- 

 ness to procumbent. In size the differences are very remarkable, the 

 extremes being as 25 to 1. In such characteristics as the secretion of 

 oils, tannin, resin, etc., there are also striking differences — ^in some 

 varieties one or more of these being present in a pronounced degree 

 and in others almost entirel}^ absent. Such a character as the v/inter 

 condition of the leaves affords opportunity for conspicuous variations. 

 The leaves ordinarily are quite persistent through the winter, giving 

 the plant every appearance of an evergreen, but some forms tend to 

 drop many of their leaves in the winter and become partially deciduous, 

 and the extreme is reached in at least one case, which is absolutely 

 deciduous. 



As the plants are pollinated by minute insects crawKng about in 

 the flowers, it was naturally expected that cross-polhnation would be 

 of frequent occurrence and that many of the dift^erent forms would 

 prove to be of hybrid origin and would follow the usual custom of 

 hybrids in the segregation of characters. This complete permanency, 

 however, of all the varieties argues against the idea of their hybrid 

 origin. Assuming them to be mutants, we would expect them to breed 

 true, as they do. Pure cultures of practically all the varieties are being 

 maintained and any mutations that may occur will be observed. So 

 far none have been recorded. It is quite improbable that the plant 

 has reached the end of its capacity to originate permanent variations, 

 and although all of the hundred or more varieties now existing originated 

 previous to their being brought under cultivation, we can look with 

 some confidence to at least some continuation of this process under 

 observation. 



All attempts at cross-pollination between different varieties have 

 thus far given no results. That they are, at least in a large measure, 

 sterile to one another is evident. To what extent hybrids are possible, 

 and between what varieties, and their behavior, are questions yet to be 

 determined. 



The Role oj the Factors in a Desert Complex in Evolution Processes in 

 Leptinotarsa, by W. L. Tower. 



During the present year, substantial structures (of reinforced con- 

 crete and cypress) have replaced the temporary ones hitherto used. 

 In addition to the new arrangements for the actual experiments, a 

 complete system of water-pipes supplies each cage, saving much time 

 and Vv'astage of water and in other respects contributing to the efficiency 



