296 SCIENCE PROGRESS 



is one of the least suitable plants for genetic investigation, 

 and any university working with it would require heavy 

 subsidies, which in their turn would demand results. When 

 each plant requires a square metre of ground, when every 

 flower used for seed has to be artificially prevented from 

 crossing by bees, and when all the field work must be carried 

 out in a sub-tropical climate, the genus cannot be regarded as 

 a convenient one for the purpose, even if we disregard the 

 trouble which ginning involves in the handling of pedigree seed. 



The author had the privilege to experience several years of 

 such work under the Khedivial Agricultural Society of Egypt, 

 during which time his efforts were in part directed to ana- 

 lysing a few crosses between Upland and Egyptian cottons. 

 The first few results gave the impression that the interpre- 

 tation of cotton hybrids on Mendelian lines would be an easy 

 task. Then, as the size of the families grew, and other crosses 

 were examined, cases which had seemed simple showed them- 

 selves to be complex. Gametic reduplication, reversion on 

 crossing, and similar phenomena were found at every turn, 

 these being in no way abnormal or non-Mendelian, but each 

 in itself raising matter for a separate piece of research. The 

 task was evidently more than one worker could handle, and 

 simpler crosses between Egyptian strains were made to elucidate 

 the previous records ; these in their turn frequently showed 

 complexities of no mean order. The latter work was carried 

 on, but in order to satisfy economic demands which had 

 arisen in the meanwhile, a simpler attack, and application of data 

 previously obtained, was made by the isolation of pure lines. 

 At the end of nine years' work the first deliberate synthesis of a 

 desired type of cotton had just been accomplished, in one family 

 only. 



It is difficult under these circumstances to see what 

 possibility there is of rapid development in the study ot 

 Genetics in Cotton. The maintenance cost is too great for any 

 university, in proportion to the results obtained ; the time and 

 skill required are too great to make such work a paying- 

 proposition for any commercial body ; the immediate urgencies 

 of the economic situation can be satisfied by the simpler work 

 of preparing and testing pure strains from existing stocks by 

 a refined rule of thumb. 



The prospects of science in this direction are thus not very 





