CERATA OF DENDRONOTUS. 351 



increasiDg per centages of alcohol rising up to about 75%. 

 Afterwards stained "in toto " in picrocarmine, treated 

 with acidulated alcohol, dehydrated, embedded in paraffin 

 in the usual way, and cut with the Cambridge "Rocking" 

 microtome. 



In order to make the sections of a manageable size, I 

 cut off the anterior and posterior moities, leaving the 

 central portion bearing the first and second pairs of cerata 

 only. A study of the sections showed me that they had 

 not been cut quite horizontal, the anterior end of the 

 section being relatively higher than the posterior. I 

 traced, first of all, the anterior prolongations from the main 

 lobe of the liver, w^hich w^e had figured* running towards 

 the first pair of cerata. These I could follow through 

 several sections, lying alongside the stomach, to a point 

 immediately below the first pair of cerata. But I found 

 as T advanced through the sections and gradually rose to 

 a higher level, that the liver processes disappeared, and 

 disappeared before the stomach ; which showed that they 

 terminated actually before the level of the top of the 

 stomach was reached. I then directed my attention to 

 the second pair of cerata. I picked up the main mass of 

 the liver at a point which I estimated to be immediately 

 below them, and as I advanced through the sections, 

 gradually rising to a higher level, I traced the formation 

 of two lateral portions distinct from the central mass 

 (PL XII., fig. 1, I'). These lateral portions were evidently 

 the lateral caeca, previously figured! as running up towards 

 the bases of the second pair of cerata. But I now find 

 that these lateral masses actually disappear before the 

 central mass ; the right hand side one, two sections beyond 

 the one figured (fig. 1), and the left hand side one, four 



*Loc. cit., ?1. XII., fig. 1. 

 tioc. aY., PL XII., fig. 1. 



