452 Dr. J. E. Gray on the Classification of Sponges. 



and others are trifid. In other genera the spine is short, and 

 the lobe is slender and weak and forked at the end, which 

 gradually pass into spicules which have the lobes variously 

 divided into branches in a most unequal and irregular man- 

 ner, gradually passing into others which have an orbicular 

 horizontal disk at the end of the short spine instead of the 

 lobes or hooks. 



This order presents the greatest abundance of spicules and 

 the most diversified forms of them. The spicules that form 

 the greater part of the skeleton of these sponges are most fre- 

 quently united together by an extra development of siliceous 

 substance. Dr. Bowerbank has repeatedly denied that the 

 latter is a true explanation of their structure, and calls them 

 siliceo-fihrous sponges. Any one who will grind down any of 

 the siliceous network of these sponges, so as to expose 

 their internal substance, will see the perfect form of the spi- 

 cules, and the additional deposit of siliceous matter which 

 unites them together. This deposit is formed of thin concen- 

 tric coats, like the spicules. The same thing may be seen by 

 submitting a similar piece of the skeleton of the sponge to the 

 action of a spirit-lamp, when the different layers of the 

 cementing portion and spicules separate. This structure is 

 well shown in Prof. Claus's beautiful work on Euplectella. 



Dr. Bowerbank, in the ' Proceedings of the Zoological So- 

 ciety,' 1869, pp. 66 & 323, has pubhshed " a Monograph " of 

 the ^' Siliceo-fibrous Sponges," illustrated with eight plates 

 by Lens Aldous. I have the utmost confidence that these 

 plates accurately represent the specimens in the slides placed 

 before the artist ; but knowing how many of the specimens 

 so mounted were obtained and manipulated, I have great 

 doubt of the fragments figured belonging to or fairly repre- 

 senting the structure of the species they are said to illustrate ; 

 at least, I know that they are taken from very different parts 

 of the sponges. Thus what is figured as Myliusia Grayu 

 was a very minute fragment which was nipped off from the 

 upper margin of a minute sponge, about the size of a large 

 thimble ; and that which was described as DactylocalyxPrattii 

 is from a specimen cut from the expanded root of the sponge. 

 Now it has never been proved that the structure of two such 

 different parts of a sponge is identical, and therefore that frag- 

 ments, taken from different parts, fairly represent the generic 

 or even specific character of a sponge. The specimen which is 

 described as the type of the genus Myliusia of Bowerbank, as 

 distinct from my genus of that name, is taken from a very 

 young and imperfectly developed sponge which, I believe, 

 belongs to a very large species. It is to my mind very doubtful 



