540 MR. G. A. BOULENGER ON [DeC. 7, 



The examples from which tliis species is described measure 

 respectively 25-10, 26*10, and 28-30 inches, and were all taken in 

 Port Jackson, where this species is locally known as the "Drummer" \ 

 I have never seen a small specimen of this fish. 



Breeding : only one, the largest of my specimens, showed any 

 signs of breeding ; this was a male with the milt but little developed ; 

 all three were taken during the month of August. 



As food: not held in any estimation, and commanding no sale in 

 the market, nevertheless it is, in my opinion , quite equal to the other 

 herbivorous Sparoids. 



Habits : in these it is a true rock-fish, dwelling in the crevices 

 and indentations of our rocky shores, where it finds abundant food 

 and shelter ; it is not given to roaming, and is only taken by the 

 trammel, one end of which is attached to the shore, against which 

 the mesh must actually lie, or else the fish would assuredly pass 

 inside, whence it happens that this species is almost always caught 

 within a few feet of the shore. 



Note. — From Dr. Ramsay's MS. notes on Australian Fishes, I 

 find that, so far back as 1881, he noticed these differences with regard 

 to the dentition, but never published any communication thereon. 



5. On the South- African Tortoises allied to Tesiudo 

 geometrica. By G. A. Boulenger. 



[Keceived November 2, 1886.] 



(Plates LVII. & LVIII.) 



Upon the suggestion of the Rev. Mr. Fisk, of Cape Town, who 

 has enriched the Society's Menagerie with so many interesting 

 Reptiles, I have undertaken a reexamination of the South-African 

 Tortoises allied to Testudo geometrica, and am able to distinguish 

 as many as seven well-marked species, of which the diagnoses follow. 

 The specimens named T. trimeni, after the Director of the South- 

 African Museum, and T. JisJci, were lately exhibited in the Society's 

 Gardens, and were unrepresented in the Natural History Museum. 

 That named T. smithi, after the author of the ' Illustrations of 

 South-African Zoology,' is established on a specimen erroneously 

 referred by Gray to T, verreauxii. The true T. verreauxii being still 

 unrepresented in our collections, its diagnosis is compiled from Smith's 

 description and figure. 



^ It sh.area the name with Girella elevata, Macleay, and Pachymetopon grande, 

 Gunth. In the ' Annals and Magazine of Natural History ' for November 1886, 

 Dr. Giiuther described Pimclepterus sydneyanus, n. sp., from Port Jackson, and 

 suggested that Pachymetopon grande (Cat. Austr. Fish. i. p. 106) may be Pime- 

 lepterus fuscus, Lacepede, and that Pachymetopon squamosum, Macleay and 

 Alleyne (^Proc. Linn. N. S. Wales, i. p. 275, pi. ix. f. 1), may bo Pimcleptems 

 cinerascens, Forsk., or P. tahmel, RiippeU. 



