446 BULLETIN OF THE 



surface of the visceral mass into the upper chamber above the septum. All 

 were turgid; some had already burst, and partly discharged their contents into 

 the chamber; others seemed on the point of doing so; the alcohol had coagu- 

 lated the escaping ova in situ, in the most perfect manner, the whole process 

 thus being displayed. It is probable, as suggested by me in 1886, that the 

 chamber serves to some extent as a marsupium or shelter for the ova and 

 young, and that they are not discharged into the surrounding element at once. 

 This is undoubtedly the case in Modiolarca. 



Verticordiid^e. V. aculicostata (I. p. 285). Another specimen, and a re- 

 examination of the one reported on in 1886, confirm the description then 

 given. There are no palpi, the anterior pair are wholly unrepresented, the 

 posterior or lower pair may be represented by two small rounded hardly 

 elevated tubercles between the mouth and the anterior ends of the gills. The 

 foot is relatively extremely large, round, and stopper-like. The gills in the 

 second specimen are clearly adnate, as in Pelseneer's figure of Lyonsiella papy- 

 racea Smith (pi. iii. fig. 1), except that they are underlaid by the solid 

 fleshy siphonal septum, and do not as in Lyonsiella serve to supplement that 

 septum. They are proportionately very much smaller, hardly reaching behind 

 the middle of the foot. I suspect that the free end of the gill in my first 

 specimen was separated by a lesion, and is not normal, but that the gill is 

 always adnate in the adult condition. 



The septum is thick and fleshy, quite destitute of perforations or orifices 

 except that in which the foot stands. 



Verticordia tomato, Jeffreys, by an accident, was not struck out of the list of 

 Verticordice (I. p. 286) after I had determined it to belong to Poromya 

 (I. p. 281). 



The balance of characters will perhaps carry Mytilimeria and Lyonsiella to 

 the Anatinidce, or a family by themselves, rather than to the Verticordiidm, 

 where I placed them. But they are transitional in their relations, and in 

 spite of the relations between the form of the gills in Lyonsiella and Lyonsia 

 I am still inclined to think the former almost equally close to Verticordia. 

 The discrepancy noted by Pelseneer arises from the fact, that, instead of com- 

 paring Lyonsiella with a genuine Verticordia like acuticostata, as I did, he 

 compares it with a species of Poromya, which is, of course, a very different 



thing. 



Poromyidje (I. p. 280). In 1886 I separated from Poromya the forms 

 which, when adult, have the hinge teeth obsolete, under the name of Ceto- 

 concha. This group included not merely those with a double posterior row of 

 modified septal orifices on each side, such as C. bidla, the type, and C. mar- 

 garita, but also certain species of Poromya, in which the hinge teeth are feeble 

 or obsolete in the adult, while in the typical Poromya they continue strong. 

 I called attention to the fact that the soft parts of these species did not differ 

 essentially from Poromya (I. p. 280), but hardly felt justified in separating 

 them from the typical Cefoconcha;. It is probable that it would be better for 

 them to form a section of Poromya, which may be called Cetomya ; while the 



