Development of the Marine Sponges. 333 



to right, as indicated by tlic arrows respectively on the illus- 

 tration (fig. 21, /?, i), and as if the cilia on its surface were 

 arranged spirally. 



When the embryo is crushed it is found to be filled with 

 sarcode charged with cells and granules of different sizes, 

 together with the spicules of the species (figs. 21 & 22, ^), the 

 latter very delicate, and the larger cells filled with smaller 

 ones (figs. 21 & 22, c), as if they were the commencement of 

 the ampullaceous sacs, which we shall by-and-by find so nu- 

 merous in the perfected sponge. 



While the cells of the coloured mass (fig. 22, a) at the poste- 

 rior extremity are of the same size throughout, and independent 

 of the colouring-matter of the sarcode which surrounds them, 

 the cells of the ectoderm of the body and of the coloarel mass 

 are, respectively, about i, 3, and 2-6000ths inch in diameter. 



It has already been stated that the ciliated ectoderm and 

 the spicules make their appearance while yet the ovum retains 

 its spheroidal sha])e ; and it might be added that all the forms 

 of the spicules belonging to the species make their appearance 

 about the same time, whereby, even at this early stage of de- 

 velopment, the species may thus be determined to which the 

 ovule belonged, as will be seen by the illustration of an 

 ovum of Esperia cegagropila^ Cart., var., on the rocks here 

 (fig. 25), wherein the skeleton-spicule {h) and the three forms 

 of flesh-spicules, viz. anchorate (c), bihamate (J), and tri- 

 curvate (e), are all present. The tricurvate here is very long 

 and straight, as may be seen by fig. 26, which represents one 

 in its mother cell from the parent sponge, thus chiefly consti- 

 tuting the " variety." 



It should be particularly remembered, however, that the spi- 

 cules in the embryo are not confusedly dispersed throughout 

 the body-substance, but are, in their natural position, confined 

 to i\\Q ptosterior part, close to the root-cells, where the long ones 

 are grouped parallel to each other and to the longitudinal 

 diameter of the embryo, with their large ends posterior and 

 their small or pointed ones anterior (Plate XXII. fig. 28, e). It 

 is only when the embryo is carelessly crushed on the slide, 

 for microscopical examination Avitli a high power, that they 

 appear to be generally dispersed throughout the body. 



Having placed twenty embryos of Halichondria simulanSj 

 such as that above described, in a glass vessel (finger-glass) 

 two thirds filled with fresh sea-water, I observed that while 

 they often attached themselves, by the posterior end, to the 

 bottom of the vessel, some became shorter in lengtli, with a 

 corresponding widening (Plate XXI. fig. 22) ; and thinking tliat 

 they wanted to become fixed, while remembering the habit of 



Ann.& Mag. N. Ilist. Ser. 4. Volxix. ' 24 



