SAGITTA GAZELLAE 2 47 



Johnston & Taylor assumed that the description ' claw '-shaped applied both to the tip and to the base 

 of the hook, and thus the hooks of their specimens seemed to have lyra and gazellae characteristics; 

 it is clear from Ritter-Zahony's report that the description ' claw shaped ' applies only to the base of 

 the hooks, and so, in this respect, Johnston & Taylor's specimens were S. gazellae. 



Nevertheless, these authors had reason to doubt the validity of the shape of the base of the hooks as 

 diagnostic of 5. lyra owing to Ritter-Zahony's inclusion of S. furcata Steinhaus in the synonymy of 

 S. lyra. One of the diagnostic features of S. furcata was the forked tail, and this is a feature which 

 occurs in many species (possibly owing to damage). Undoubtedly many specimens of S. furcata were 

 S. lyra, but it seems likely that others were not. Thus, the hook of S. furcata (Fowler, 1905, pi. 4, 

 fig. 12) (Fig. 3) referred to by Johnston & Taylor as being of the simple gazellae type (yet from a 

 species synonymous with 5. lyra), may well have been taken from the specimen shown in Fowler 

 (1905, fig. 7, p. 14) (Fig. 4) which appears to be S. hexaptera; or alternatively it may have come from 



Fig. 1. 



Fig- 3- 



Fig. 4. 



Fig. 1. A 'typical' hook from S. gazellae (a) and a 'secondary' hook from S. lyra (b). (From Ritter-Zahony, 191 1.) 



Fig. 2. The tip of a hook of S. 'lyra'. (From Johnston & Taylor, 1921.) 



Fig. 3. A hook of S. furcata'. (From Fowler, 1905.) Fig. 4. A specimen of S. furcata. (From Fowler, 1905.) 



a young specimen of S. lyra and still be correct. Neither the existence of claw tips in S. gazellae hooks 

 nor the presence of simple type hooks in certain specimens of S. lyra affect the validity of Ritter- 

 Zahony's statements. 



It is upon Johnston's & Taylor's evidence that the case put forward by Tokioka (1939) and Thomson 

 (1947) for the inclusion of S. gazellae in the synonymy of S. lyra mainly depends, and this evidence 

 has been wrongly interpreted. 



Hamon (1952) has shown that there is no second generation of hooks (the 'secondary' hooks of 

 Ritter-Zahony) in S. lyra, but that in sexually mature specimens the existing hooks become ' claw '- 

 shaped. She has suggested that this type of hook is a secondary sexual character. 



In S. gazellae a claw-type hook is found in the specimens at stages III, IV and V of maturity, but 

 the claw shape is not nearly so pronounced as in S. lyra. There is no sign of a ' claw ' in the inner 

 margin of the hook, but only in the thin plate at the base, whereas both the inner margin and the base 

 are 'clawed' in S. lyra. Comparison of Fig. $a-c demonstrates this point. 



The typical claw-type hook of S. lyra is characteristic and remains a useful feature in differentiating 

 the species from S. gazellae. 



