IMMATURE SPECIMENS OF PENELLA FILOSA. 303 



As with P. fihsa, the horns of P. varians develop as a pair of 

 outgrowths from the sides of the posterior part of the cephalo- 

 thorax. The metamorphosis of the anterior part of this region 

 of the body to form the " head " is also fundamentally similar in 

 both species. In detail, the development shows noticeable dif- 

 ferences. With P. filosa no division of the thorax into segments 

 is traceable in my youngest larva, although this has no horns or 

 " head." Later ; the sides of the anterior cephalothorax become 

 inflated, early destroying the convexity and distinct borders of 

 the region. The mouthparts are gradually replaced by tuber- 

 cules, the first of which to appear arise in a definite order. With 

 P. varians (Fig. 5) the sides of the anterior region of the cephalo- 

 thorax beneath the lateral borders of the shield early form two 

 rounded outgrowths, while nearer the posterior end of this part 

 of the body the horns protrude as two longer, pointed outgrowths. 

 This occurs before the separate thoracic segments lose their 

 identity and while the cephalothoracic shield is still distinct in all 

 parts. In an older specimen of this species, the anterior or 

 "head" outgrowths are shown to be broadened and have fim- 

 briated margins. At this time the boundaries between the 

 cephalothorax and thorax and those for the individual thoracic 

 segments are no longer distinguishable while the borders of the 

 cephalothoracic shield are still recognizable anteriorly and the 

 mouthparts are intact. The later metamorphosis is not known. 



The rostrum, mouthparts and antennae of the P. varians larvae 

 seem to have been very like those described for my larvae. The 

 thoracic limbs are figured with the two anterior pairs bi-ramous, 

 the two posterior pairs uni-ramous. No account is given of the 

 changes which the cephalothoracic appendages undergo, nor of 

 the development of the lateral appendages of the abdomen. It 

 is clear that the latter were not present in any of the larvae de- 

 scribed and the implication is that their development is markedly 

 retarded, although no definite statement is made with regard to 

 this point. It is not improbable that in this species the adult 

 shape of body and " head " may be attained before the abdominal 

 appendages are developed. Such would appear to be the case 

 with another species of Penclla at least, P. c.voccdi, for Steenstrup 

 and Liitken ('61) figure specimens which have a "head" of 



