AND THE MOEPHOLOGY OF THE EYE IN INSECTS. 393 



lioptron has the form of a round or oval truncated cone, of which the base and apex are 

 iuhspheroidal or, more generally, cycloidal surfaces*. The older writers named the 

 nflected ring which unites the cornea and the basilar membrane "t7ie sclerotic" ; I shall 

 lesignate it the scleral ring. 



The inner margin of the scleral ring is often thickened, and receives the insertion of 

 me or more muscles (figs. 1 & la. mc), which are probably concerned in the adjustment 

 accommodation ?) of the optical apparatus. 



The whole interior of the dioptron is divided more or less perfectly into hexagonal 

 >r square tubes by fine sheaths f which extend from the margins of the corneal 

 acets to the basilar membrane ; each encloses a cone and a great rod, and forms a 

 segment of the dioptron. 



The great eyes of the dragon flies (JEshtiia and Agrion) have enabled me to detect 

 arge lymph-sinuses (a.b. figs. 5 & 9) which jtass through the scleral ring and connect the 

 ymph-spaces of the dioptron with those of the head. 



In JEshnia the afferent vessel (a) enters the superior internal border of the dioptron : it 

 las distinct walls, and is the anterior extremity of the bifurcated aorta %. I am unable 

 :o say that this is general in insects, but from what I have seen I suspect that it is. 



The efferent openings (b.b. figs. 5 & 6) are situated in the lower portion of the scleral 

 ring, nearly opposite to the afferent vessel : they are generally, even in very different 

 insects, long slits, across which a number of radiating nucleated fibres rf are stretched, 

 between the basilar membrane and the scleral ring. These in sections, vertical to the 

 basilar membrane, present a fan-like arrangement similar to that of the ciliary muscle 

 of the vertebrate. 



I have been unable to make out any stride in these fibres ; and as non-striated muscles 

 are not known to occur in insects, I hesitate to regard them as contractile, although one is 

 almost tempted to such a belief from their arrangement and position ; perhaps they are 

 simply elastic bands which maintain the tension of basilar membrane, where its attach- 

 ment to the scleral ring is defective, to permit the circulating fluid to pass out of the 

 dioptron. Their occurrence, however, in the simple eye (fig. 33, Is.) around the edge of 

 the corresponding membrane is indicative of an active function. 



The tracheal vessels (t.v.) of the dioptron consist of a main trunk, which almost surrounds 

 the edge of the basilar membrane, from which numerous branches ramify and anasto- 

 mose on the neural surface of the membrane (fig. 68). 



Vessels from this network perforate the basilar membrane and run outwards, almost to 

 the cornea, between the cuticular sheaths of the segments of the dioptron ; they termi- 

 nate in blind extremities. 



In the SyrphidcB the tracheae of the dioptron assume the form of fusiform sacs 

 between its segments ; these have very narrow necks where they pass through the basilar 

 membrane. 



The basilar membrane is a cuticular structure, which was described by Leydig as a 



* Phil. Trans. I.e. p. 595. t " Umhiilluiigsschlauche " of Leydig & authors. 



t ' Anatomy of the Blowfly,' by the author, plate is. fig. 1. 



