52 



small size of the stomach, oesophagus and recliim. the more conspicuous 

 since the body is fairly elontjate. The arms aic slill (|uile siiorl : the anterioi- 

 edge of the oral lobe is lather thickened, the [)reoral vibralile i)and 

 being not yet very distinctly difl'erentialed. The skelelon is very chaiacl- 

 eristic, the elongated body rods being rather coarsely thorny anfl ending 

 in Iwo serrated branches*) (Fig. 15); I lie poslorai rod is fenestrated. 

 There is a small recurrent rod, with the end curving forward, and the body 



skeleton accordingly 

 does not assume a 

 basket structure. 



After 8 days the 

 larvae had reached the 

 shape shown in PI. X, 

 Fig. 6. The postero- 



dorsal arms have 

 developed, their rods 

 being fenestrated. All 

 the arms are fairly 

 broad, but somewhat 

 abruptly narrowed to- 

 wards the j)oint. 

 There is a small pro- 

 minence on each side 

 at the anterior end, indicating the place of the future preoral arms, 

 and further there is a characteristic small thickening on the postoral 

 band where it bends out along the postoral arm. Vibratile lobes or 

 epaulets have not yet been formed, and it then remains undecidetl 

 whether they exist in this larva; Judging, however, from the presence of 

 epaulets in the MespiliaAarvA as also in the undetermined Temnopleurid 

 larvai described below it can scarcely be doubted that epaulets, but no 

 vibratile lobes, will be found also in this larva. (Cf. the larva from the 

 (iulf of Siam, mentioned below.) There are only scattered pigment cells, 

 no prominent pigment spot in the point of the arms. 



In the skeleton (Fig. 16) there is still only little change. The absorption 

 of the body rods has begun, one of the end branches having disappeared. 



') In the preliminar> iioliic "On (lie duvelopnienl of some Japanese Efliinodernis" p. .346, 

 il is stated that the main rod of the body skeli'lon in Ihc Tcnmopk-urid Iarv;v "lias a niecUan 

 process off Ihe posterior end of tlie stomach, and l)ey(itid tliis process tlicre is a sliorl iiroion"- 

 ation. whicli becomes alisorbed in the later stages". While il really has this ai)pcarance when 

 seen from the ventral or dorsal side, the side view of the body makes it evident that it is more 

 correct to say that the body rod ends in two branches, one of which (tlie ventral) may be 

 directed more or less directly downwards: this latter appears to be the first to be absorbed, 

 when the full larval shape is about to be formed. 



Fig. l.j. Skeleton of larva of Temnopleurus loreumalicus 

 stage. A. from the ventral side; B. side view. ^""/i. 

 Letters as in fig. 3. vtr. ventral transverse rod. 



1st 



