494 TRANSACTIONS OF SECTION D. 
aster appears, which forms the skeleton for the club-shaped aboral spike which 
now grows out. This skeleton consists of a lattice of three diverging rods with 
cross-connections. The apex of the club is covered with a pad of thickened epithe- 
lium carrying cilia. The base of the club skeleton forms connections with the 
body and recurrent rods. So far as I can make out, both right and left coelomic 
vesicles send up prolongations which become madreporic pore-canals, and open 
to the exterior by right and left madreporic pores. Subsequently these pores 
merge in a single median pore, from which right and left pore-canals diverge. 
As the larva grows in size, other arms sprout out. The postero-dorsal arms, as 
in other Echinoidea, receive their skeleton (which is latticed) from a new cal- 
careous aster on each side. The preoral arms, which are in front of the mouth, 
central to the antero-lateral arms, receive their support from branches of a single 
median aster situated above the cesophagus. So far eight arms are developed, 
as in other Echinoid larve. But in Echinocardium other arms make their appear- 
ance. Two of these, the antero-dorsal, receive their skeleton as branches of the 
rods supporting the preoral arms : two others, the postero-lateral, are supported 
by outgrowths from the base of the club skeleton. As these arms grow in length 
the aboral club undergoes absorption and is reduced at the time of metamor- 
phosis to a short rounded prominence. 
As in other Echinoid larve an invagination appears on the left side, the floor 
of which becomes converted into the oral disc of the young Spatangoid. On this 
disc the adult spines grow and in the centre the adult mouth is formed. The 
disc increases in size till it takes in all the left side and part of the right side of 
the larva, displacing the anus to one side. Then the larval arms are rapidly 
absorbed : the stomodwum becomes disconnected with the gut and shallows out, 
and the oral portion of the larva shrinks to an insignificant knob. The young 
Spatangoid has no spines on the aboral surface (in this it contrasts with the young 
Kchinus). All its spines are pointed except a circle situated at one end; these 
have broad flattened ends and constitute the only distinctive Spatangoid feature 
as yet visible. The mouth is surrounded by five plates, from each of which 
springs an inwardly directed spine. From the rhythmical movements of these 
spines I am inclined to think that they represent the teeth of regular Echinoids. 
3. A Preliminary Account of the Development of the Starfish Asterias 
rubens (L.). By J. F. Gemmmn, M.D. 
Artificial fertilisation at the Millport Marine Biological Station; complete 
subsequent development in Embryology Laboratory, Glasgow University ; methods 
employed to secure circulation, &c., in the hatching vessels. 
General chronology of development until completion of metamorphosis; move- 
ments of the larve; the coelomic fluid circulation; a pulsating ‘ heart-complex.’ 
Mode of development of the chief coelomic cavities; probable homologies of 
these cavities on the right and left sides; light thrown on this question by larvae 
with double hydroceele; comparison with Balanoglossus. 
4. On the Hybridisation of Echinoids. By H. M. Fucus, B.A, 
The work described was done in the laboratory of the Marine Biological 
Association, Plymouth, in conjunction with Dr. Cresswell Shearer and Mr. 
W. de Morgan, during 1909-12. The present paper deals chiefly with the experi- 
ments of this spring and summer. An account of the work of 1909-11 has already 
been published. ! 
In order to arrive at definite conclusions from experiments on the hybridisa- 
tion of Echinoderms two principles must be followed : (1) The hybrids should be 
healthy. The maldevelopment of a character in an unhealthy larva can easily 
vitiate the results. (2) The characters of which the inheritance is investigated 
should be definite in the parental forms—preferably present in the one while 
absent in the other. The variation curves of the pair of characters should not 
overlap. Owing to the elaboration of methods of rearing Echinoids up to and 
1 Jour. M.B.A., vol. ix. 
