656 Mr. R. Slielford on the Pterylosis of the 



the embryo ; being all directed backwards they constitute a 

 regular flowing mane (Plate XIII. A), the exact boundaries 

 of which are not easy to define. None of the actual defini- 

 tive feathers have yet made their appearance, so that the 

 following description applies strictly to the distribution of 

 the trichoptiles. 



Ft. capitis (fig. 1 ^, p. 655^ pt. cap.). — This is very well 

 developed on the crown and back of the head, but is 

 sparse on the sides ; the skin below the ear, between 

 the mandibular rami and between the eye and nostril is 

 naked, a few delicate threads are to be seen surrounding 

 the upper and posterior borders of the ear-opening and 

 on the gonys of the mandible. The eyelashes have not 

 yet made their appearance. 



Pt. spinalis (fig. 1, pt. sp.) is confluent with the pt. capitis ; 

 at first it is single, but at about the level of the attach- 

 ment of the humerus it bifurcates, the two branches 

 again reunite in the lower lumbar region and run on to 

 the pygidium, stopping short just in front of the oil- 

 gland papilla. The apterium between these two spinal 

 branches is very narrow, and might easily be overlooked 

 had one no later stages at hand for comparison. 



Pt. humeralis (fig. \,pt. hum.) arises from the pt. spinalis just 

 below the point where the bifurcation begins, and runs up 

 to the point of the shoulder to fuse with the trichoptiles 

 on the patagial membrane ; its connection with the pt. 

 ventralis is not yet established, nor does the ill-developed 

 parapteron join it. 



Pt. femoralis (fig. 1, pt.f.) is a triangular tract, the long 

 trichoptiles converging from the back over the whole of 

 the outside of the thigh to a point at the knee ; anteriorly 

 the base of this triangle is confluent with the pt. spinalis 

 for a short distance, but it soon diverges and runs down 

 as far as the anterior angle of the pygidium, leaving 



* The figure representing Stage 2 is introduced here for expediency, 

 since the difference between it and Stage 1 is ahnost imperceptible 

 (see p. 658). 



