96 Mr. W. Marshall on the 



may break up into any number of parts, and then these, if 

 they are not to disturb the equilibrium, must always collect 

 in such a manner that every plane drawn longitudinally 

 through the creature may divide it into two lialves of equal 

 weight and of the same structure. If the divisional parts 

 are all of equal size, the mass of cells, except when it is of a 

 conical form, has no anterior or posterior, superior or inferior, 

 part founded on structure (although perhaps on movement) 

 {Blastula). It is otherwise, however, when the divisional 

 parts are of different sizes and of different weights, and are not 

 sufficiently numerous for one set of them to group themselves 

 round the other as a continuous mantle ; then the larger ones, 

 in order to preserve the equilibrium, as also the smaller ones, 

 will arrange themselves in a particular manner [amphihlastula 

 of Sycon), and in this way special regions of the body will 

 orientate themselves. That at the same time a cavity may 

 be formed by separation in the interior of the regularly con- 

 structed cell-aggregate, and that this may become filled with 

 water, is of no consequence if only the parts of the wall 

 remain in equilibrium. Matters remain the same if this 

 cavity breaks through outwards, or if a portion of the hollow 

 sphere becomes invaginated ; whether a swimming-gastrula 

 forms a central cavity by invagination or by perforation, its 

 mouth must be placed centrally and the parts of the wall 

 must arrange themselves around it and the primary stomachal 

 space, so that here also the equilibrium remains intact. So 

 soon as special organs become differentiated in our animal, even 

 without its acquiring a definite permanent position of the direc- 

 tion of movement necessitating the distinction of upper and 

 lower, right and left, and whether these organs are such as aid in 

 the taking of food (tentacles), or complications of the digestive 

 cavity (mesenterial sacs, gastral canals), or sexual organs, &c. 

 &c., the central place being already occupied by the primor- 

 dial stomachal space, these must always place themselves 

 peripherally, occur in plurality, and group themselves in such 

 a manner that the animal does not lose its equilibrium. In 

 this way the radiate structure of the primitive swimming 

 Ccelenterata was brought about. There is, however, another 

 point that must not be overlooked in these considerations. A 

 radiate structure is not only of great service, indeed to a 

 certain extent indispensable, to free-swimming creatures of 

 spherical or cylindrical form &c., on account of the stability 

 of equilibrium, but it may also be of great importance to 

 sessile forms, inasmuch as, acting externally, it harmoniously 

 increases the power of resistance in all directions ; we con- 

 etruct not only our air-balloons on the radiate type, but also 



