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



Family 8. Phakelliadce. The author, in endeavouring to vary the 

 mode in which I have described the genus Phakellia, states that the 

 " skeleton is formed of closely reticulated horny fibres, forming an 

 expanded mass," &c. This is incorrect, as it is not a kerato-fibrous 

 sponge, but belongs to the order Silicea. See Mon. Brit. Sponges, 

 vol. i. p. 186. 



Family 10. PoJymastiadce, p. 503. "Sponge with tubular fis- 

 tulous branches ; tubes open at the end, and formed of longitudi- 

 nal and transverse fascicules of fibres." 



Here, again, the author has fallen into a great error in describing 

 the "fistulous branches" as open at the end. In no species of 

 Polymastia that I have ever seen is that the case. See Mon. Brit. 

 Sponges, vol. ii. p. 73. 



Family 12. Evplectelladee, p. 504. "Skeleton composed of 

 longitudinal, transverse, and oblique bundles of spicules intersecting 

 each other, and forming a network." This description of the struc- 

 ture of the skeleton of Eiiplectella, Owen, as I have before stated, 

 is incorrect, the whole skeleton being truly siliceo-fibrous. 



Family 13. Esperiadce, p. 504. "Sponge massive. Skeleton 

 composed of fusiform and linear spicules, interspersed with aucho- 

 rate, bihamate, or birotulate spicules. Sarcode soft." 



The description of this family is singularly incomprehensible. 

 " Sponge massive." This character would apply to by far the 

 greater number of sponges in existence. "Skeleton composed of 

 fusiform and linear spicules." All the four primary forms of skeleton- 

 spicula are liable to fnsiformity ; which of the four forms does the 

 author mean, any one in particular or all of them ? And when he 

 names "linear," what form of linear spiculum does he mean? as 

 all straight or slightly curved spicula are linear. " Interspersed with 

 anchorate, bihamate, or birotulate spicules." This embraces a most 

 extensive field of investigation, as we shall presently see. " Sarcode 

 soft." All sponge-sarcode is soft. Thus it will be seen that every 

 one of the characters assigned by Dr. Gray to the Esperiadce are 

 singularly indefinite. Subsequently, at page 532, he divides the 

 family into six sections, which are about as ill-defined as the charnc- 

 ters of the fiimily. In the first place, the author designates all the 

 retentive spicula as defensive spicula. The defensive spicula, exter- 

 nal and internal, are quite another class of spicula than those peculiar 

 to the sarcode. See Mon. Brit. Spong. vol. i. p. 21. 



The application of the retentive, or bihamate and anchorate, spi- 

 cula to the distinction of families or genera is peculiarly unfortu- 

 nate. I will not trace their occurrence through the exotic genera 

 of sponges, but confine myself to their range among the British 

 genera. They are met with in Microciona, Hymedesmia, Hijme- 

 niacidon, Halichondria, Isodictya, and Desmacidon — six genera, con- 

 taining 126 species; and of this number of species, 51 only have 

 their spicula imbedded in their sarcode ; and it rarely occurs that a 

 species has only one form, and frequently three forms are present 

 mixed together. 



Family 14. Tefhyadce, p. 504. The author describes the ske- 



