32 



There has been much difference of opinion as to the validity of the genus Diphyphyllum 

 of Lonsdale, chiefly arising from the fact that its original' founder erroneously made its sup- 

 posed fissiparous mode of division, a generic character, Milne Edwards and Hainie, how- 

 ever, pointed out that the mode of increase in the corals referred here is truly by gemmation, 

 and not by fission. Hence, they refused to accept the genus, and placed the corals previously 

 referred here under Cyathophyllum and Lithostrotion. Upon the whole, however, there can be 

 little hesitation in adopting the view advocated by Mr. Billings, (Can. Journal, new series, 

 Vol. IV. p. 134,) and in retaining Diphyphyllum as a distinct genus. 



Diphyphyllum, as above defined, differs from Cyathophyllum in its distinctly bi-areal 

 structure, the septa not extending to the centre, as they do in the latter genus, nor being 

 twisted together, so as to produce the appearance of a small columella. 



From Lithostrotion, which it closely resembles in most respects, Diphyphyllum is separated 

 by the invariable absence of a columella. Milne Edwards and Haime, (" British Fossil Corals," 

 p. 195,) express their opinion that the absence of the columella in Diphyphyllum is accidental 

 and due to the process of fossilisation. A.S pointed out by Mr Billings, however, there can be 

 no question as to the incorrectness of this opinion, the columella being entirely wanting in the 

 most beautifully preserved examples of Diphyphyllum. 



From Amplexus, Diphyphyllum is separated by the general form of the coral, by the 

 greater development of the septa, and by the absence of a septal fossule. 



In all essential points of their structure the genera Eridophyllum and Diphyphyllum are 

 identical, the sole difference of importance being that the corallites in the former genus are 

 united to one another laterally by horizontal root-like prolongations of the epitheca. It must 

 be admitted, however, that specimens of Diphyphyllum occasionally exhibit similar connecting 

 processes, though these are never developed to any extent, and can only be detected by a care- 

 ful examination. 



The rharacters of the genus Diplophyllum (Pal. N. Y., Vol. II. p. 115),founded by Hall 

 for the reception of a common coral from the Niagara Limestone, do not appear to differ 

 essentially from those of Diphyphyllum, except that the outer vesicular layer in the former is 

 stated to be separated from the central tabulate area by a distinct mural investment. It is 

 probable, however, that Mr. Billings is correct in uniting the two genera. 



Two species of Diphyphyllum have been described by Mr. Billings from the Corniferous 

 Limestone of Western Ontario, being amongst the most conspicuous corals of the formation. 

 I have also detected specimens inseparable from D. gracile of McCoy, if indeed the latter is 

 not truly identical with D. stramineum (Billings). 



26. DEPHYPHYLLUM ARUNDINACEUM (Billings). 

 (Plate VI. Fig. 1.) 



Diphyphyllum arundinaceum (Billings), Canadian Journal, New Series, Vol. IV, p. 134. 



"Corallum forming large masses of long cylindrical straight or flexuous stems, from 

 three to four lines in diameter, but usually distant from one to three lines from each other; 

 radiating septa thin, betwen forty and fifty in number, rarely reaching the centre; transverse 

 diaphragms turning downwards on approaching the margin ; two to four in one line. In some 

 of the corallites the walls are so thin and closely united that no separation can be observed, but 

 in others of the same cluster an outer area is distinctly .visible. There is usually a circular 

 space in the centre of the corallites, into which the radi tting septa do not penetrate ; often 

 however they reach the centre. The young corallites sometimes spring from the sides 

 of the parent with a slender base, and curving upwards immediately become parallel with those 

 of the whole group. In large colonies frequent instances may be seen, where, instead of this 

 lateral budding, a bifurcation takes place, both branches being of the same size. In large 

 groups, owing to the numerous additions of young, the corallites diverge slightly, as if radiat- 

 ing from a point. The colonies are from six inches to several feet in diameter, and large 

 blocks of stone are of frequent occurrence, which are penetrated at right angles to the stratifi- 

 cation by the closely crowded stems " (Billings). 



Diphyphyllum arundinaceum (Billings) is distinguished from the following species 

 chiefly by the almost uniformly greater size of the corallites, which average three or four lines 

 in diameter when fully grown. In shape the corallites are cylindrical, with annulations and 



