MttLLEB] THE) AIR-SACS OF THE riGEON 381 



the interclavicular sac itself. Beyond, it extends backwards past the 

 oval diverticulum of the saccus cervicalis, into the space between the 

 scapula and the anterior ribs, where it attains its greatest extent 

 (figs. 2, 11, and 12, DSSC3). Its posterior end usually lies upon 

 the second true rib under the scapula. Its anterior end reaches as 

 far as the diverticulum oesophageo-tracheale. Near the oval diver- 

 ticulum of the cervical sac it communicates with the latter by means 

 of a small but sharply defined triangular opening. Sometimes, but 

 not always, I have seen it communicate also with the suprahumeral 

 diverticulum. In these cases a delicate canal passed outward along 

 the lower border of the scapula and the musculus scapulo-humeralis 

 anterior, and opened into the suprahumeral diverticulum near the 

 tnberculum inferius humeri. 



It should also be mentioned that in two cases the principal portion 

 of the diverticulum subscapulare was separated from the inter- 

 clavicular sac and its own proximal portion by a partition which, 

 in the region of the brachial plexus, divided the diverticulum 

 into two chambers. In these cases the subscapular diverticula 

 were, of course, always connected with the suprahumeral diver- 

 ticulum, and each of the three chambers appeared to give rise 

 to an anterior diverticulum. The communication between the sub- 

 scapular and suprahumeral is not infrequently present, and by no 

 means restricted to the cases where the principal portion of the 

 subscapular diverticulum is divided from the interclavicular sac. 

 But still, although not uncommon, this is exceptional, and I was 

 much surprised to find this state of things described as the normal 

 condition by Roche, who says (1891, p. 31) : "Ces deux derniers 

 (saccules sousscapulaire et humeral) ont une origine commune, 

 sortant du reservoir claviculaire par un orifice menage en arriere du 

 muscle petit abducteur de lnumerus." Baer (1896, p. 436) ex- 

 presses himself similarly. I have not found in the literature on the 

 air-sacs of the pigeon any statement concerning a connection between 

 the diverticulum subscapulare and the saccus cervicalis, but it has 

 been described in the domestic fowl. 



From the ventral side of the diverticulum subscapulare an out- 

 growth (figs. 2, 11, and 12, DSSC3) arises, which extends, parallel 

 to the second false rib, towards the processus lateralis anterior 

 sterni, without, however, reaching it. It is often broken up into a 

 row of small saccules, and may be so wide at its base that the diver- 

 ticulum from which it arises appears drawn out locally, as it were, to 

 form it. In this case the outgrowth attains a quite unusual size. 

 The walls of the diverticulum for the most part adhere to the sur- 

 rounding muscles, only a small portion being free. 



