399 



tracheal trunks, but in the Attidae by a mere rudiment in the form 

 of a short lateral process on each side at the base of the two large trunks. 

 In the Attidae, therefore, by far the greater part of the tracheal 

 system is nothing else but a pair of modified, ectodermal tendons, 

 which are serially homologous with the entapophyses connected with 

 the pulmonary appendages. In the Dysderidae the entapophyses do 

 not form a part of the tracheae, which mast, therefore, be here con- 

 sidered as the homologa of the lungs alone. In a Tetrapneumonous 

 Spider I find the four entapophyses quite separate from the four lungs, 

 and provided with four stigmata distinct from the spiracles, — a con- 

 dition which must be considered as the more primitive. 



The remaining appendages also give rise to entapophyses, to 

 which the posterior endosternite in the adult is attached. 



Genital Ducts. — Into the cavity of each abdominal appendage 

 a hollow process from the corresponding coelomic sack is protruded. 

 Just where the medial side of this coelomic process passes over into the 

 wall of the sack itself a fold of the mesoderm is formed, which grows 

 in a lateral direction, cutting off the medial part of the intra-appen- 

 dicular coelomic cavity in the form of a short tube. This tube, which 

 is shewn in section in Fig. 2, ends blind at its medial end, while late- 

 rally it widens funnel-shaped and opens into the coelomic sack. At 

 first the tubes are of equal size in all the appendages, but later on those 

 of the pulmonary appendages develope vigorously, so that their open 

 ends grow some distance into the abdomen and finally attach them- 

 selves to the anterior ends of the two chords of genital cells, which 

 have grown forwards from their point of origin at the posterior end of 

 the germinal band. The blind ends of this pair of tubes, which gives 

 rise to the genital ducts, approach one another, growing along the 

 inner surface of the hypodermis towards the medial line, and finally 

 communicate by means of a single, transverse cleft with the ectodermal 

 genital opening. 



The tubes of the three remaining appendages become rudimentary. 

 Their presence tends to shew that the genital ducts had originally 

 some other function, and the similarity of their development with that 

 of the coxal glands in Arachnids generally indicates their nephridial 

 origin. The genital ducts in the Scorpions have recently been shewn 

 by Brauer 2 to be also of nephridial origin. 



It remains only to be noticed that the genital segment in the 

 Aranema is the eigth post-oral (or second abdominal) segment, not the 

 seventh as previously supposed, and this brings the segmentation of 



2 loc. cit. 



