64 l!KCOia»S OF THK Ai:sTKAI,lAN MrsEl'M. 



male and female upon dissectiou ; in the male the teeth were all tricuspid, 

 while they were trenchant in the female, thus bearing out the observations 

 recorded above. 



Girella hiacl-li, Castelnau, was included in the synonymy of G. tricns- 

 ^idata by Ogilby in 1893, while the identity of G. j^ercoide;:. Hector, and 

 0. simplex was recognised by Hector in 1877. Ctenolahrus Jcuo.vi, Hutton, 

 was recognised as a synonym of G. simplex by Gill in 1893. 



A specimen in the old collection of the Australian Museum is labelled 

 as Girella ranisai/i, though it unfortunately has no other data. It is 

 apparently the holotype of Macleay's species, no other example so labelled 

 being in either the Macleay Museum or the Australian Museum collections. 

 It agrees with the brief description of (t. raiimnyi in its length and major 

 details, though some proportions of the eye, snout and interorbital space 

 are a little different. It agrees with the descri[)tion in having only two 

 perfect anal spines, but the malformed base of a third is quite distinct. 

 It does not differ in any details from specimens of G. tricuspidata of the 

 same size. 



Mr. Ogilby has compared the holotype of Girella mentalis, De Vis, in 

 the Queensland Museum with an example of G. fricu><pidata, and tinds no 

 difference between them except that the holotype has only thirteen dor.'^al 

 spines. He notes that though he has examined scores of specimens of 

 Girella from Moi'eton Bay, he has not found anothei' specimen to agret' 

 with De Vis' example in this detail, and therefore regards the latter 

 mei'ely as an unusual variation of G. trifitxpid((ta. 



The type of G. carhonaria, De Vis, is apparently lost, 'i'liis imper- 

 fectly described species is here i-egarded as probably synonymons with tlie 

 female form of G. tricuspidata. 



Two male examples about 350 mm. long, recently secured at Kawau 

 Island, New Zealand, by Mr. C. Hedley, agree well with Clarke's descrip- 

 tion and figure of (Jirella multilineata. A careful coniparision of them 

 with specimens of G. fricuspiditfa from Sydnev, howevei', reveals no 

 distinguishing chai"acters to substantiate the New Zealand species. 



Distribntion. — Girella triruspidata was originally described from Shark 

 Bay, Western Australia. The species has since been recoided from 

 Victoria, Tasmania, Now South Wales, Moreton Bay, and New Zealand. 



(IiUKl.i.A KLKVAIA, ^[arleuy. 

 Blnck Drummer; Hook Blaoklish. 



( I'hito xiii.. Hy. I.) 



Girella elevata, Macleay, Proc. Linn. Soc. N.S.Wales, v., 1881, p. 408. 

 Id., Ogilby, Cat. Fish. N.S.Wales, 1886, p. 18. Id., Waite, Mem. 

 N.S.Wales Nat. Club, No. 2, IPOI. p. -jn. i,l^ Stead. Ed. Fish 

 N.S. Wales, 1908, p. 51, pi. xx. 



D. xiii/14; A. iii/11: P. 18; V. i/5 ; C. 15. 5 1 rows of scales below 

 the lateral line between the RoapnlH and tho liypurni joint and about the 



