332 



DISCOVERY REPORTS 



the loss of four and two spines, respectively. Of these ninety-one fifth furcilias ten moult to sixth 

 furcilias, whereas eighty-one miss out this three-spined stage and moult directly to the one-spined 

 seventh furcilia. The nineteen sixth furcilias that have moulted from both fourth and fifth furcilias 

 also moult to seventh furcilias. There are, then, three alternative routes from fourth to seventh 

 furcilias, 4-5-6-7 ; 4-6-7 ; 4-5-7, and it is the last of these that the majority of larvae follow. 



Table 8. Variation in the number of terminal spines on the telson 

 lost during ecdysis of the fourth and fifth furcilias 



No. of 



F7 



100 



Fig. 14. The paths of development of one hundred fourth furcilias calculated from the percentages of fourth furcilias moulting 

 to fifth or sixth furciUas and the percentages of fifth furcilias moulting to sixth or seventh furcilias (see Table 8). The figures 

 show the arrangements of the spines on the telson in each stage and it can be seen how the number of terminal spines in the 

 succeeding stage is determined. 



Although it is now accepted that, within one species, larvae with certain combinations of setose 

 and non-setose pleopods tend to be dominant and that the few other combinations that occur can be 

 considered as variants, it is unusual for any of the furcilias which are recognized by the number of 

 terminal spines on the telson, to be omitted. This point is emphasized by Boden (1955, p. 375) who, 



