310 NEW YORK STATE MUSEUM 



tracheae by osmosis. Sanderson's conclusions, to state them 

 in another way, were that the paired dorsal cavities [pl.28, fig. 

 4:pd€] are connected with the exterior at the point ew, and that 

 the wedge-shaped apical portion of the caudal spines can be 

 shut down tightly on either side so as to close this opening, 

 and that the air finds its way from the paired dorsal 

 cavities through a tube into the paired ventral cavities and 

 thence is carried along this cavity to the tracheal system. 

 From my own studies I have been led to conclude that there is 

 never a natural opening leading into the paired dorsal cavities, 

 and that, when there is such an opening present, it is an artifact. 

 The cuticle at this point is thin and easily broken, while the 

 cavity is lined with hypodermal cells; and I do not know of 

 such a condition existing anywhere among the Hexapoda where 

 there is a cavity opening to the exterior and lined by an epithe- 

 lium. It is true that the mouth and rectal openings are lined 

 with epithelium, but in both of these cases the lumen end of 

 the cells is covered by a well marked cuticle. In addition to 

 the above, I have found that the wedge-shaped portion is always 

 rigid and not capable of any motion, while any air that might 

 find its way into the paired ventral cavities, if an opening did 

 exist, would be unable to reach the tracheal system, because 

 the base of this cavity is completely plugged up by a portion 

 of the cuticular pocket, as has already been described [pl.28, 

 fig.8j<; and fig.Gprc]. I have been unable to find any tubes such 

 as Mr Sanderson describes, while his so called elliptical open- 

 ings are not openings but areas in the cuticle, that are very thin 

 and trans])arent and very liable to lead one into error if one 

 worked only with thick free-hand sections. When the caudal 

 spines are mounted in Canada balsam, so that their dorsal 

 surface can be studied, two rows of these elliptical openings can 

 be seen forming the octal surface of apparently well marked 

 grooves or furrows, but these grooves or furrows are notliiug 

 more than the cavities of the paired dorsal cavities into which 

 one looks through the transparent outer cuticle. This is further 

 confirmed by the study of cross sections of the caudal spines; 



