374 DR ARTHUR T. MASTERMAN ON THE 



apertures of the ovaries to the exterior are, however, situated upon the oral surface of 

 the animal, an adaptation to the habit of ' carrying' the eggs which was conjectured by 

 Sars. A series of sections through the parent at once demonstrates the small inter- 

 radial openings just at the edge of the oral disc. As the ovary becomes ripe the eggs 

 gradually increase in size and their protoplasm becomes filled with an enormous quantity 

 of small yolk-granules of a pale yellow hue. These granules, as far as I can ascertain, 

 are deposited in the substance of each ovum by intra-cellular secretion, and are not pro- 

 duced by the disintegration of adjacent ova nor by the secretory activity of follicular 

 cells. 



Brooding Habits. — The phenomena of oviposition and of fertilisation have escaped 

 my observation, and the present account commences with the fertilised eggs and their 

 subsequent history. Each female lays from thirty to fifty eggs, which are found in the 

 so-called ' brood-chamber ' ; Sars found from ten to thirty. The female attaches itself 

 by the tube-feet of the distal half of each arm to the surface of the rock, and then bends 

 the proximal half of each arm at right angles upwards, at the same time arching its oral 

 disc. In this way is produced a bell-shaped brood-cavity, into which the eggs are shed 

 and retained. In this position the mother starfishes are found adhering to the perpen- 

 dicular face of rocks, or more commonly hanging downwards under overhanging ledges. 

 They do not occur very far above low-tide-mark, but large numbers are found completely 

 uncovered by the tide. In every case they appear to be capable of retaining water in 

 the brood-cavity, the necessity for which may account for this predilection for hanging 

 downwards. So far, the habits are very similar in Asterias miilleri, but there are slight 

 differences to be noted. The habit of ' brooding ' is not so intense in Cribrella, and the 

 slightest disturbance at any stage serves as a pretext for the female to forsake her duties. 

 It is, however, fair to add that the forsaken eggs or larvse do not appear to suffer by 

 such treatment, and proceed in their development • in a similar way to those still re- 

 tained. In Asterias miilleri the maternal instincts are stronger, and the parent will 

 retain her eggs in confinement under considerable vicissitudes. Similarly, the eggs and 

 larvse of the former are never at any time mechanically attached to the parent nor to 

 each other, but in the latter each larva attaches itself firmly to its ruptured egg-mem- 

 brane, and the whole forms a solid ball of larvse, which is firmly retained in situ by the 

 parent. So much is this the case that the late larvse can only be separated at great 

 risk of mechanical injury, whereas in Cribrella the embryos can be washed out by a 

 pipette. During the period of ' brooding ' the parent can obviously obtain no food, and 

 in consequence the internal organs undergo some remarkable changes which cannot be 

 here described. If the eggs of a Cribrella be removed no amount of force will serve to 

 straighten out the arms, breakage always taking place, but if the animal be left undis- 

 turbed on its aboral surface it will in the course of about half-an-hour or so assume a 

 perfectly normal shape, and shortly after will greedily devour any food presented to it. 

 I have never yet succeeded in coaxing a female to re-commence her maternal duties 

 after the eggs have once been washed out of her brood-cavity. If a Cribrella fixes 





