2O2 KATHARINE FOOT AND E. C. STROBELL. 



All such deviations from a method of development that has 

 been claimed to be sufficiently rigid to justify far-reaching 

 generalizations are a warning against formulating general laws 

 that do not take into account all the facts. 



If our results demonstrated in Anasa tristis be shorn of all 

 their significance except their admission as normal variations, 

 they still stand as a protest against premature generalizing. 

 We expressed this in an earlier paper (1911) in reference to our 

 demonstration that the accessory chromosome in Anasa tristis 

 does not always fail to divide in the second spindle. 'If the 

 significance holds which has been attributed to the failure of this 

 chromosome to divide in one of the two maturation divisions, 

 then those cases in which it divides twice in Anasa must be set 

 aside as pathological, the spermatozoa resulting from these 

 divisions having no functional activity. Only to those who 

 have endowed the chromosomes w T ith causal attributes can 

 variations in their behavior cause embarrassment. If we find 

 them subject to marked variations it is a characteristic they 

 have in common with other structures in the cell the nucleolus, 

 the centrosome, the mitochondria, the polar rings and other 

 structures which have been found to be so variable that interest 

 in speculations as to their possible causal significance has steadily 

 waned, and those who believe the chromosomes are equally 

 variable may justly suspect that the hypotheses surrounding 

 these structures may be destined to the same fate as the specu- 

 lations so long surrounding other cell organs, notably the centro- 



some." 



Nearly all cytologists agree that experimental breeding has 

 been and may ever be the most trustworthy test of hypotheses 

 which are based on the morphology of the cell and which claim 

 to offer a mechanical explanation of heredity. 



The chromosomes of the Hemiptera are largely responsible for 

 the modern chromosome hypotheses, for certain stages in their 

 growth and development which are assumed to justify these 

 hypotheses can be clearly followed and demonstrated. Those 

 Hemiptera therefore which show the chromosome phenomena 

 necessary to a given hypothesis are especially fitted to test its 

 value by experimental breeding. 



