822 SECTIONAL TRANSACTIONS.—D. 
after the worms had separated, contained no cocoons. In its essential features the 
process of coition follows that already described for L. terrestris (Q.J.M.S., vol. 69, 
pt. 2), and the improved technique rendered it possible to make direct observation 
of the fact that the flow takes place on both sides of each worm, making in all four 
streams of seminal fluid. 
By isolating worms it has been possible to establish the fact that the solitary 
worm produces cocoons. The act of cocoon deposition has been watched and has 
been found to be quite separate from that of coition. It has also been found that the 
time which elapses after the cocoon has left the clitellum until it is quite free from the 
worm is so short (from three to five minutes), and the movement is so continuous, that 
the usually accepted view that the cocoon receives the eggs and sperm whilst it is 
assing over the apertures of the oviducts and spermathece, respectively, is improb- 
able. This has been confirmed by the fact that when the worm has been killed at the 
moment when the cocoon was ready to leave the clitellum, but was still surrounding 
it, eggs were present in the cocoon. - 
Saturday, August 29! 
Cruise on the Solent. 
Monday, August 31. 
Mornine. 
Representatives of Section D took part in a discussion on Biology as an 
Element in the Science Curriculum of Schools. (See p. 376.) 
17. Dr. W. K. Spencer.—The Evolution of Starfish. 
Starfish are not found in any of the known earliest fossil-bearing rocks (Cambrian), 
but suddenly appear in fair numbers and in great variety of type in the Ordovician of 
Europe and N. America. The forms which then appear include only a few species 
which are strictly comparable to the starfish as ordinarily defined and there are no 
true brittle-stars. 
They can be placed in groups, if physiological, rather than morphological, characters 
be adopted in classification. In other words, the variety of morphological type can 
be understood best from considerations of progression, capture of food and digestion 
of food. 
The course of evolution appears to be as follows: The earliest Kchinodermata 
seem to have fed on small food, which they seized whilst retaining a stationary 
(possibly sessile) position. The great change to freely moving forms followed a change 
ot habit, namely, the chase of ‘ large ’ food. 
Not any new organs are developed in consequence of the change of habit, but the 
existing skeleton is modified to suit the new conditions. The ossicles in the mouth 
region are arranged to form a food trap and, as the animal can now indulge more 
freely its own taste in food, it elaborates methods of chase and powers of perception. 
Last in order of evolution come improved methods of the digestion of ‘ large ’ food. 
18. Dr. G. Kinestry Nosie.—The Relation of Life-history to Phylogeny 
within the Amphibia. 
Unusual modes of development have frequently been described within the 
Amphibia, but no attempt has previously been made to trace out a phylogeny of these 
life-histories. Recent advances in our knowledge of the anatomy and natural group- 
ings of the Amphibia have permitted a re-examination of these data from the standpoint 
of the genetic relationships of the genera. A distinctive type of life-history was found 
to run through each natural group of genera considered, showing that the mode of 
development is frequently more constant than many adult characters. A close 
agreement was found between life-history and recent classifications based upon the 
anatomy of adult material. A progressive change in life-history may be traced with 
some accuracy in the Urodeles. In the Salientia, yolk size has changed so frequently 
