Br. Tf. /. Barkas — Hyhodus, a Coal-measure Fish. 163 



transversely ; those of contiguous rows being usually opposite or 

 nearly so. Dissepiments non-poriferous. Four (sometimes three) 

 alternating rows of cells on each branch, opening by minute rounded 

 pores, which are prominently elevated above the general surface. 

 Five pores to the length of a fenestrule. Eeverse unknown. 



Polyzora iuherculata differs from all the other recorded species 

 of the genus, except P. venucosa, McCoy, in the possession of pro- 

 minent-edged pores, which give to the celluliferous surface of the 

 branches a peculiar tuberculated or pustulose appearance. The 

 latter species, however, has exceedingly long and narrow fenestrules, 

 with nearly twice as many pores to the length of a fenestrule. I 

 have only seen fragmentary examples of this species. 



Locality and Formation. — Eare in the Hamilton Group, Bartlett's 

 Mills, near Arkona, Township of Bosanquet. 



Eetepora Phillipsi, Nicholson. PI. IX. Fig. 21. 



Polyzoary infundibuliform, or forming a flattened expansion. 

 Branches slender, fifteen to eighteen in a quarter of an inch 

 measured transversely, parallel, flexuous, not united by dissepiments, 

 but coalescing at intervals of about a third of a line, and over 

 spaces of the same length. Fenestrules oval, six or seven in the 

 space of a quarter of an inch measured vertically, seven or eight 

 in the same space measured diagonally (five in two lines on an 

 average), regularly alternate in contiguous rows. Cells in two 

 alternating rows upon each branch, the rows separated in decorti- 

 cated specimens by a distinct impressed line. Three cells to the 

 length of a fenestrule, sometimes only two ; and one or two opposite 

 the points where the branches coalesce. Eeverse unknown. 



This is a genuine Betepora, and in its general form and its biserial 

 cells is closely allied to B. prisca, Goldfuss, which I have found 

 abundantly in the Corniferous Limestone of Ontario. It is, how- 

 ever, readily distinguished by the more slender, rounded, and ap- 

 parently non-angulated branches, the much smaller size of the 

 fenestrules, and the greater number of these openings in a given 

 space. I have named the species in honour of Professor Phillips, 

 to whom we owe so many descriptions of Devonian and Carbonifer- 

 ous fossils. 



Locality and Formation. — Corniferous Limestone, Port Colborne. 



(To be concluded in our next Number.) 



*^* Plate IX. will accompany the concluding part of this paper in our next Number. 

 — Edit. Geol. Mag. 



IV. — Hybodu^, a CoAii Measure Fish. 

 By W. J. Barkas, L.R.C.P. Lond. ; M.R.C.S. Eng. 



THEEE appear to be two or three obstacles in the way of a proper 

 classification of fossil fishes, and of obtaining sufficient data to 

 establish the earliest period cf the earth's formation, during the 

 deposition of which any particular fish existed, the principal obstruc- 

 tion being caused by the disconnected manner in which the researches 

 have been carried on ; for paleontologists, as a rule, confine them- 



