128 



Journal New York Entomological Society. t'^'o'- ^^- 



This represents a great variation. In fact every space on the 

 ocular micrometer scale between .29 mm. and 1.16 mm. is shown in 

 the table. 



From the figures in the two tables there are two things which 

 stand out clearly: the fact of the variation in the number of moults, 

 and the fact that a larva may moult, and yet not increase in size, or 

 with very little increase. 



Considering first the matter of variation it may be shown that this 

 may be reduced to some order. I have said that the ratio of increase 

 of the head widths was practically .35. The average measurements, 

 made as given in the large table, are here compared to the theoretical 

 figures at that ratio. 



Average. 



Stage I 33 



n 51 



III 67 



IV 85 



V 1.04 



VI 1. 13 



Theoretical. 

 •35 

 •47 



•63 



•8s 

 1. 14 

 1. 14 



In giving the average for stage Vl measurements not included in 

 the table are used. The average was made from a separate set of 

 larvae. Stages V and VI may be compared to one figure in the theo- 

 retical column, since in the ultimate stage of certain saw-fly larv?e 

 (in which the larv.x do not feed) the head width is practically the 

 same as in the stage preceding it. In using the term " ultimate " I am 

 following Dyar (1896). In this ultimate stage the larva is orange 

 in color, and clean and dry of all slime. These orange larvje enter the 

 soil soon after moulting and are seldom seen. 



Doctor Dyar (1895) has recorded some measurements for the 

 head widths of the stages. These are: I, .25 (?) (not measured) ; 



