258 



DISCOVERY REPORTS 



of the Stage distinguished by having five terminal spines on the telson is small compared 

 with those of the stages before and after it. The individuals of E. vallentini with five 

 terminal spines on the telson do not as in the other species make a natural group : some 

 of them have an unaltered antenna and are in other ways nearest to, or indistinguishable 

 from, those with five pairs of setose pleopods and seven terminal spines ; the remainder 

 have the antennal endopod segmented, the exopod scale-like, are bigger, and have the 

 thoracic limbs and gills more developed. For this reason the Furcilia larvae of E. 

 vallentini with five terminal spines on the telson are divided between two stages. 



In the group of stages recognized by the number and character of the pleopods 

 E. longirostris and E. spinifera have one stage more than E. frigida, E. vallentini and 

 E. triacantha ; in the group recognized by the number of terminal spines on the telson 

 E. triacantha has one stage more than any of the other species. This is shown in Table II. 



Table II. Showing the characters of the Furcilia stages in the five species of 

 Euphausia and the numbers by which they are known 



The table shows that a given stage of one species is not necessarily the same as the 

 same stage of another species : stage I of E. frigida has four pairs, stage I of E. longi- 

 rostris one pair, of non-setose pleopods. The same set of characters may in two species 

 be diagnostic of two different stages : a Furcilia with four pairs of setose and one pair 

 of non-setose pleopods belongs if it is E. frigida to stage II, if £. spinifera to stage III ; 

 a Furcilia with three terminal spines on the telson belongs if it is E. vallentitii to stage V, 

 if E. triacantha to stage VI. 



The stages recognized by the number of terminal spines on the telson are not so well 

 defined as those recognized by the character and number of the pleopods. The range of 

 size is bigger among the former so that they overlap more than do the latter. This is 

 shown diagrammatically in Fig. 35 for the three species of which I saw large numbers of 

 larvae. The columns show the range of the lengths of the individuals of each stage ; the 



