Appendages in Hexapoda. 207 



pression of the structure, due to the bending of the abdomen, 

 which is angularly applied to the thorax. 



Organs which we may regard as homologous with these 

 ventral sacs are met with first among the Chilopoda in the 

 genera Lithobius and Henicops, standing still nearer to the 

 Protosymphyla, in the coxse of the last four, or more rarely 

 five, pairs of ambulatory legs, where they occur as thread- 

 spinning coxal glands. In the Scolopendridse and Geophi- 

 lidse, derived by elongation from the shorter primitive forms, 

 analogous organs, here characterized as pleural glands (on 

 account of the union of the coxse with the pleura;), occur 

 in the last leg-bearing segment. 



Among the Symphyla a lobiform plate appears in Scolo- 

 pendrella immaculate/, on the coxse of the second pair of legs, 

 and this in the next segment is transformed into a ventral 

 sac which is only slightly protrusible. The distal part is 

 covered with a transparent homogeneous chitinous cuticle, 

 and lined with a few gland-like hypodermic cells. Below 

 this layer of cells lies the reticulated tissue of the adipose 

 mass, through which blood- corpuscles pass into the ventral 

 sac. Outside of this coxal saccule, as we must call it here, 

 there is to the thirteenth segment a claw-like appendage 

 increasing in size posteriorly, which can by no means be 

 regarded as the rudiment of a leg, but only as the product of 

 transformation of a joint-spur, and which occurs similarly on 

 the two posterior pairs of coxse in Machilis *. On the twelfth 

 segment the coxal sac is reduced to a softer, oval, mem- 

 branous piece ; in the undeveloped legs of young animals we 

 find no trace of appendages on the coxse. 



In the order Diplopoda also protrusible sacs situated in 

 the coxse are often present ; thus they occur in the anterior 

 segments in Chordeumidse and Lysiopetalum, as well as in 

 the section of the Colobognatha derivable from the Chilo- 

 gnatha, in Polyzonium and Siphonophora, and indeed they 

 appear first, and at the same time most strongly developed, 

 on the third pair of legs, the somite of which would therefore 

 correspond to the first abdominal segment of the Hexapoda, 



As these ventral sacs in the coxse of the Myriopoda, or at 

 the posterior margins of the ventral plates of the Thysanura, 

 usually occur at the end of partially unconnected develop- 

 mental series, we are compelled to assume their probably 

 polyphyletic development within the order. And yet, in 

 their position, in their origin, and at the same time in their 



* To what extent such structures, originally equivalent to the ordinary 

 cutaneous seta?, can become developed, is shown especially by the tibial 

 spur, e. (j. of the Ileterocera. 



