On the Perilymphatic Spaces of the Amphibian Ear. 243 



The Saccus perilymphaticus in the larvae of R. fusca and Pelo- 

 bates, has at certain stages a peculiar connection with the tracheal 

 chamber, oi* the lung of its own side (figs. 21, 21, st.). In Pelobates 

 the Saccus is bounded ventrally by a thick membrane, continuous as 

 already stated, with the outer perichondrium of the chondrocranium. 

 Closely attached to this, at a short distance posterior to the foramen 

 perilymphaticum inferius, is a dense fibrous strand, luhich passes ver- 

 tically downivards to become connected ivith the dorsal ivall of the 

 lung (fig. 21). The elements of which the strand is composed are 

 chiefly elongated cells, bearing a considerable resemblance to non- 

 striped muscle fibres. A similar arrangement is found in R. fusca tad- 

 poles (fig. 27). From a comparison of various developmental stages of 

 the latter, the fibrous strand appears to be a specialized portion of a 

 longitudinal suspensory ligament, by which at earlier stages the lung 

 is suspended partly from the floor of the skull. The ligament does 

 not consist of especially dense tissue, and whilst its longitudinal line 

 of attachment to the lung is narrow, its dorsal attachment is broad, 

 and extends from the base of the skull outwards to the soft tissues 

 lying below the dorsal aorta of its side. With the exception of the 

 dense fibrous strand — or column, as it might almost be called in 

 Pelobates — the ligament almost completely disappears as develop- 

 ment proceeds, nothing remaining of its connection Avith the floor of the 

 skull, except the strand in question. This represents a growth of fib- 

 rous tissue, such as was in nowise foreshadowed by the degree of de- 

 velopment reached by the longitudinal ligament, of which it an hyper- 

 trophied localized remnant. Near its attachment to the lung, it is 

 connected with a flat band of fibres running horizontally below the 

 pharynx (in Pelobates) to become attached in a precisely similar way 

 to the corresponding region of the opposite side (fig. 21, .s^'). This 

 band has a considerable antero-posterior extent and is attached to each 

 lung along a line running longitudinally on its dorsal surface. It 

 need not however be further discussed here. In Rana fusca, as will 

 be seen from fig. 27, the fibrous connection above described is from 

 the floor of the saccus to the wall ot the tracheal chamber, on each 



side, and not to the lungs as in Pelobates. 



16* 



