308 PROCEEDINGS OF THE ACADEMY OF [May, 



age of its segment, but close to the spinnerets (Sp.) at the posterior 

 edge of the third segment. The stigma (T. St.) is a slight transverse 

 groove, the trachea (TV.) itself is a conical plug of ectoblast cells push- 

 ing into the archiccelic space between hypodermis and mesoblast 

 (Mes.). Successive later stages of the ectoblastic impaired ingrowth 

 (T. St.) are shown in median section in figs. 13-15; in these stages 

 it already shows the division into an outer vestibulum, transversely 

 widened, from the middle of which extends forward a single tracheal 

 trunk, all with a thin chitinous lining. Fig. 15 is the stage of hatching, 

 aged fifteen days, and shows just behind the stigma the still solid 

 primordium of the colulus (Col.). 



Thus the appendages of the third abdominal segment disappear 

 entirely by merging with the circumjacent hypodermis and without 

 forming any invagination. Quite independent of them and at the 

 posterior margin of the third abdominal segment arises subsequently 

 an unpaired ectoblastic inpushing, the vestibulum, from the middle 

 part of which grows forward a little before the time of hatching a single 

 tracheal trunk. I have not found any indications of folding of the 

 tracheal wall like the pulmonary folding, such as Simmons described, 

 and I would judge that the fold supposed by Simmons to be the trachea, 

 represented in his fig. 8, is not a tracheal invagination at all, but merely 

 the space dorso-posterior to the caudal lobe. 



Reference may be made, in passing, to the relative position of the 

 segments in the adult abdomen. The pulmonary stigma marks the 

 posterior border of the second segment, the tracheal stigma (or, in 

 Dipneumonids, that of the second pair of lungs) the posterior border 

 of the third segment, while the anterior and posterior spinnerets mark 

 the posterior regions of the fourth and fifth segments respectively. 

 In Loxosceles the third segment comes to occupy a large portion of the 

 venter, as shown in figs. 22-24, PI. XIII. In Filistata (fig. 33, PI. XIV) 

 and Evagrus (figs. 30-32) the third segment is relatively much smaller 

 After the stage of reversion the first abdominal segment cannot be 

 distinguished externally from the second, and I am inclined to think 

 it enters mostly into the composition of the pedicel. 



3. Development of the Lung-books. 



A considerable amount of work has been done upon the development 

 of these organs since Salensky (1871) showed that they are derived 

 from the appendages of the second abdominal segment. 



Bertkau (1872) found that new lamellae are added during the growth 

 of the spider, each new one as an invagination from the region of the 

 lateral corner of the stigma. 



