Ixxii 
THE VOYAGE OF H.M.S. CHALLENGER. 
The calthrops may have been derived from a trisene by shortening of the rhabdome, 
or from a micro calthrops by increased growth. Possibly it has been in some cases 
evolved in one way, in others in the other ; thus in Tetilla merguiensis it probably 
arises from the protrisene by a reduction of the rhabdome ; in Pachastrella it has the 
appearance of an overgrown microcalthrops, but considering the close alliance of 
Pachastrella with Poecillastra {Normania), in which both trisenes and calthrops occur, 
the latter possibly derived from the former, this appearance may be misleading, and I 
prefer to leave the question open. 
In concluding this discussion I would point out the purely hypothetical nature of 
the view which regards the trisene as derived from the rhabdus, and there is a good 
deal to be said for an opposite hypothesis which would derive the trisene from the 
microcalthrops ; thus Poecillastra may be in a direct line of descent with Placinastrella 
{Placinastrella copiosa, Schulze), and in this sponge trisenes and calthrops, both 
apparently derived from microcalthrops, occur ; and further, it is possible though not 
probable that the dichotrisene of Thenea may have been derived from the trilophous 
microcalthrops of Placina trilopha, Schulze. In that case opposing evidence might be 
reconciled by attributing a different origin to the Tetillidse and the Theneidse. The 
spiraster in the latter family, however, would seem to preclude such a separation, if as 
seems probable this spicule has originated from a sigmaspire. If on the other hand we 
could derive the sigmaspire from the spiraster, the probability of the descent of the 
trisene from the calthrops would be enhanced, indeed the simplicity which at once follows 
the adoption of this view is so great that nothing but the stubbornness of the onto- 
logical data prevents me from adopting it. 
The Lithistid desma, since in some families it commences as a microrabd and in 
others as a microcalthrops, might be supposed to be of dual origin, and to this view I 
felt forced when first studying this group; subsequent investigation has convinced me, 
however, of the truth of Oscar Schmidt’s observations, which prove that a gradual 
transition from the tetracrepid to the monocrepid desma occurs in species of Macan- 
drewia. 
The study of the transformations of the aster will help us to understand this, 
for as already noticed we not unfrequently find it passing into a microrabd, and 
we are led to suppose that the microcalthrops which serves as the crepis of the 
tetracrepid desma has undergone in the monocrepid desma a similar reduction ; such 
embryonal variation as this would seem to imply is not unknown in other groups of 
animals. 
The transition is supposed to be from the tetracrepid to the monocrepid desma and 
not in the reverse direction, because several considerations, which will be discussed later, 
lead us to suppose that the Lithistids characterised by tetracrepid desmas were the first 
evolved, from Choristida allied most closely to the Pachastrellidse. 
