2 o DISCOVERY REPORTS 



It would be interesting to undertake the histological examination of caruncles derived from a series 

 of juvenile and adult Ceratias. This should reveal changes of structure, and possibly function, which 

 could be compared with the present inferences based on measurement only. 



THE SYSTEMATICS OF CERATIAS 

 When Bertelsen had demonstrated, by his work on the fishing lure, the invalidity of the genus Mancalias 

 Gill, and so reduced the Ceratiidae to the genera Cryptosparas and Ceratias, he reviewed the species 

 of the latter genus in a synopsis* reproduced below : 



Ceratias Kroyer 1844 

 Illicium with long slender stem. Two caruncles. 



1. Bulb of illicium in adult specimens with one filament holbolli 



(a) Dorsal fin concealed under skin, stalked caruncles sub sp. xenistius 



(b) Dorsal fin of normal structure, sessile caruncles sub sp. sessilis 



(c) Dorsal fin of normal structure, stalked caruncles sub sp. holbolli 



2. Bulb of illicium with a pair of filaments tentaculatus 



This synopsis may now be re-examined and further simplified. 



Apart from xenistius, all the forms here set down have been investigated, either as types or represen- 

 tatives, in the previous section on the growth of C. holbolli. Of 21 dimensions investigated for allometry 

 in each, the results, without exception, are consistent with the conclusion that these forms are all either 

 juveniles or adults of a single species. It is relevant here to notice an observation of de Beer (1940, 

 p. 380), who, referring to species with allometric organs, stated : ' whilst there is in such cases incon- 

 stancy of form, there is constancy of form-change and this, measured by the value of the growth constant a, 

 is also a specific character.' 



Assessing the members of Bertelsen's synopsis in order, it appears that the first, C. holbolli xenistius, 

 should remain as a subspecies for the time being. Regan & Trewavas (1932) described this form as 

 Mancalias xenistius from a single specimen. The dorsal fin was most unusual, being virtually concealed 

 and only appearing externally as (p. 99), ' a series of four pores which are the openings of black glands '. 

 Possibly this is a teratological specimen. 



The subspecies sessilis was described by Imai (1941) as Mancalias sessilis and distinguished by the 

 absence of a stalk to the caruncles. Imai gives a list of measurements, and this form appears in Table 3 

 and Fig. 5 as no. 2. There are no discrepancies. With regard to the caruncles, it is seen (Fig. 5D) that 

 their length fits the simple allometry equation for sexually immature individuals. They are pear- 

 shaped in appearance because their stalks are less constricted than those of the club-shaped caruncles 

 of other specimens. This is doubtless a minor individual variation and cannot be assigned subspecific 

 rank. 



Bertelsen's reference to ' stalked ' or ' sessile ' caruncles in his synopsis appears to be due to a confusion 

 of Imai's meaning. It has been shown in the previous section that, because of enantiometry in adult 

 life, the caruncles may become truly sessile in the ageing individual (e.g. Regan's specimen of Ceratias 

 holbolli). Previous to this late ontogenetic change, the caruncles of all members of the genus are 

 characteristically more or less pedunculate, and reference to ' stalked ' as distinct from ' sessile ' caruncles 

 is misleading in a taxonomic context. The revised species C. holbolli now consists of the two subspecies 

 xenistius and holbolli only. 



Bertelsen's second species C. tentaculatus includes Mancalias tentaculatus Norman, 1930 and 

 Mancalias bifilis Regan & Trewavas, 1932. These two species had been distinguished from each other 



* As published, Bertelsen's synopsis contained typographical errors which have been corrected by reference to his text. 



