L. DONCASTER 



broods '12.1 and '12.21, in which females were in very great excess, 

 and the remaining thirteen to broods '12.18, '12.25, '12.31 and '12.32, 

 in which males and females were in approximately equal numbers. 

 In every one of these 39 larvae, with one exception which will be 

 discussed below (brood '12.25), all the really good figures showed 55 

 chromosomes (Figs, 3 — 8). The following is an analysis of the counts ; 

 all were made by drawing the chromosomes and then counting the 

 chromosomes drawn, and in many cases the figures were drawn again 

 later and the counts compared. I have classified the figures by notes 

 made at the time into five grades of trustworthiness — 'excellent,' 'good,' 

 ' fair,' ' probable ' and ' doubtful ' — the ' doubtful ' group including those 

 in which it was not possible to decide with any confidence between 55 

 and 56. Classes 'excellent' and 'good' include cases where no reasonable 

 doubt is possible. 



The counts in the unisexual broods were as follows. Individual 

 larvae are represented with letters (A, B, BB etc.) ; the number of 

 ovarian mitoses of each class is given after the letter in words. 



Excellent 

 Total 8 



Good 

 Total 10 



Fair 

 Total 29 



'13.39 iJJ, one. 



'12.21 ? B, one. 



'12.18 ■} D, one; f F, two; f G, one. 



'12.31 iF, two. 



'13.39 ? CC, one. 



'12.1 ? fl, two; ? S, one; ? A', one. 



'12.21 ? C, one; ? I, two. 



'12.31 ? A, one. 



'12.32 ¥J, one. 



'13.39 ?B, two, ? f/, one; 1 BB, two; 



'13.39 B iF, one. 



? CC, one; ? II, one. 

 ? K, one; ? N, one. 



'12.21 ?B, two; ?F, three; ? G, three; 



'12.1 i A, one; ? D, two; ? JF, one. 



'12.18 i G, one. 



'12.31 iA, one; ? £, two; ? F, two. 



'12.32 ? B, one. 



•13.39 ? O, one. 



'13.19 ? B, one. 



'12.21 ?</, three; ^ K, one. 



'12.31 iD, one; i E, two; ? F, one. 



'12.32 i A, one; ? C, one; i N, two. 



'13.39 JL, one; ? GO, one. 



'12.21 ? G, one; ? ^V, one. 



'12.31 $ F, two. 



'12.18 ?D, one. 



In addition to the above there is one figure, in an ovary of a larva 



($ /) of brood '12.1, in which 56 appears the most probable number. 



Probable 

 Total 14 



Doubtful 

 Total 7 



