458 M. R. CLARE. 



careful examination of data on length of larval life, of pupal life, 

 etc., proved barren of results, except that a suggestive but very 

 rough correspondence between rates of oxygen consumption and 

 productivity, as measured by the number of flies appearing in 

 the vials, did come to light. A correspondence of this sort would 

 carry the implication that the metabolic rates of the pupae are 

 due to an impress set by the metabolic condition of the female 

 parent at or about the particular time the eggs are deposited, 

 and that this impress continues without impairment over the 

 period of larval life. The data at hand do little more than 

 suggest this possibility. 



EFFECTS OF INBREEDING. 



The central question around which this study was planned 

 was whether or not it is possible to find an index of inbreeding 

 in rates of metabolism. If metabolism were subject to control 

 as a result of genetic make-up, we should expect to find evidence 

 of it reflected in our results. There is not only no indication 

 that heredity plays any but the most general role in metabolism ; 

 but, on the contrary, we have seen some reason for believing 

 that rates are at the command of graduated environmental 

 influences. The irregularity in rates which we have stressed 

 renders it impossible to make satisfactory comparisons between 

 different matings within a common stock, and the propriety of 

 lumping all of the matings of a stock under a mean is very 

 questionable. However, this has been done in Table 5 in which 

 are presented means for most of the stocks. The eliminations 

 previously referred to made such serious inroads on the data 

 for the majority of the matings that Table V. must be built up 

 on an inadequate number of determinations. The figures for 

 stock G of the first period are based on as few as 3 determina- 

 tions ; and in all other cases on 4 determinations except where a 

 probable error is attached, this indicating that 5 or more deter- 

 minations were available. Deficient as it is, this table illustrates 

 the impossibility of separating inbred from non-inbred stocks on 

 the basis of rates of metabolism. 



SUMMARY. 



Several results emerge from this study which, it is believed, 

 should assist in defining the problem and, at the same time, 



