124 DR. J. S. BOWERBANK ON SPONGES. [Feb. 13, 



In his " 1st order, p. 502, Coralliospongia. Sponges hard, 

 coral-like, entirely formed of siliceous spicules anchylosed together 

 by siliceous matter into a network. Mass covered by a thin coat of 

 sarcode when alive." This description is eminently incorrect. No 

 one, I believe, ever yet saw the terminations of spicula united into a 

 network through the morbid action of anchylosis by means of siliceous 

 matter. The material by which they are naturally cemented to- 

 gether is keratode. The author does not seem to be aware that, 

 although sponge-fibres always anastomose more or less, sponge-spi- 

 cula never do so. And in the description of his first family in this 

 order, Dactylocalycidce, page 505, the author distinctly abnegates his 

 own definition of the order by describing the family thus : — " This 

 beautiful fam.ily of sponges is at once known by having the skeleton 

 formed of continuous anastomosing fibres formed of concentric lamina 

 of silica, forming a hard brittle network." 



Order II. Keratos-pongia, p. 503. In his definition of this 

 order, the author takes an extremely wide range of skeleton-struc- 

 ture, embracing the true sponges, Dysidea, Chalina, Phakellia, and 

 other genera differing widely in their structural peculiarities. 



"Subsection 2. Spicular Sponges {Spiculospongite), p. 503. 

 Sponge fleshy, more or less strengthened by fasciculated or scattered 

 siliceous spicules, the bundles behig sometimes slightly covered with 

 a thin layer of horny matter. The sarcode is generally abundant ; 

 in some few, as Euplectella, it is thin, mucilaginous, and deciduous." 

 This description is decidedly incorrect as regards Euplectella, 

 Owen, as that sponge assuredly has a siliceo-fibrous skeleton, and 

 not a spiculo-fibrous one. See Proc. Znol. Soc. 1867, p. 361. 



" Order III. Leiospongia. Sponge-spicules of only one kind, often 

 varying in size and shape in the same species." This is another 

 exceedingly vague description on which to found an order ; and we 

 may reasonably ask whether, if they vary in shape, they can be of 

 only one kind. There are very few species in which the spicules, 

 strictly speaking, would be only of one kind, and in every spiculo- 

 reticulate sponge they are liable to variations in size and shape. 

 The author has placed three families in his version of this order, but 

 in the illustration of each of them he violates his own rules. Thus, 

 in his list of illustrative species of Halic/iondria (p. 519), he in- 

 cludes H. /armaria and distorta, both of which have two forms of 

 spicula. 



The same error occurs in his family Pohjmastidce, — Polymastia 

 lulbosa and radiosa each having two forms, and P. robusta three 

 forms of spicula. 



The like error is also apparent in his family Clionidce, as, accord- 

 ing to Mr. Hancock's description of exotic and other species of 

 Clioia, several of them have at least two forms of spicula in their 

 structure. 



Order IV. Acanthospongia. This is a most expansive order, as it 

 would embrace every siliceo-spiculous family excepting those included 

 in the author's order 3. The Euplectelladce are erroneously placed 

 in this order. 



