358 Fernon L. Kellogg and R. G. Bell. 



velopment, the preponderance of males is due to the untimely 

 death of the females. 



A series of lots of ten individuals each were reared in 1903, 

 with the specific intention of testing the assumed influence of nu- 

 trition on sex determination. These lots included: (a) a lot un- 

 derfed during the whole of larval existence; (b) a lot underfed 

 during the second to fifth intermoulting periods, inclusive; (c) a 

 lot underfed during the third to fifth intermoulting periods; (d) 

 a lot underfed during the fourth and fifth intermoulting periods; 

 (e) a lot underfed during the first intermoulting period only; (f) 

 a lot underfed during the second intermoulting period only; 

 (g) a lot underfed during the third intermoulting period only; (h) 

 a lot underfed during the fourth intermoulting period only; (i) 

 a lot underfed during the fifth intermoulting period only. From 

 the rearing of such lots it was hoped to determine what influence 

 reduced rations might have on the determination of sex, and also, 

 if any, at what time in the larval life the influence was most potent. 

 A consideration of the records of the rearing of these lots at the 

 end of the season compels us to say : that the lots were much too 

 small to aftord trustworthy generalizations; that dissections of 

 the larvae at various ages reveal an unmistakable differentiation in 

 sex (indicated by gross differences in the reproductive glands) at a 

 time as early at least as the beginning of the third intermoulting 

 period, so that experimental lots c, d, g, h and i were distinctly 

 superfluous; but finally, that it may be affirmed from the meager 

 data afforded by experimental lots a, b, e and /, that the reduction 

 of the food supply (this reduction brought as near as possible to 

 a living minimum) did not produce any unmistakable results in 

 the way of an overproduction of males. 



Data of more interest are those derived from an inspection of 

 the records of the experimental rearing of various larger lots of 

 silkworms in 1901, 1902 and 1903. In 1901 the records for five 

 lots of twenty larvae each may be referred to : 



Lot I — Fed optimum food; no deaths before emergence of 

 moths ; produced 8 males, 1 2 females. 



Lot 2 — Fed optimum food; 2 deaths before maturity; produced 

 7 males, 1 1 females. 



