400 ME. B. THOMPSON LOWNE ON THE COMPOUND VISION 



condition, as the walls of the refractive structures are very thin, so that in all the speci- 

 mens which I have examined, the tubes of the spindle, which is very similar to that in 

 the flies, had collapsed. 



There are usually two principal layers of pigment in the dioptron — an outer layer, 

 which is thickest at the apices of the cones (fig. 1. fg x ), and an inner layer (pg 3 ) at the 

 inner extremities of the spindles. 



The pigment of the outer layer is generally contained in cells which form a kind of 

 iris around the inner extremity of each cone. It will he convenient to designate these 

 rings of pigment-cells "irides," in conformity with the nomenclature of the older writers. 

 There are, as a rule, five or six iris-cells around each cone. These send pigmented 

 fringes outward over the cone and inward over the great rods. In many insects 

 (Hymenoptera and Nematocerous Diptera) the outer set of fringes are replaced by rod- 

 like prolongations of the iris-cells. Each cell has a single pigmented rod, which has a 

 bright, highly refractive spherule at its outer end, immediately beneath the cornea. 

 These may be called ciliary rods (figs. 31 & 32. cb). 



I have found ciliary rods in all the Hymenoptera and Nematocerous Diptera which I 

 have examined. 



In the Dragon-flies the soft cone is surrounded by pigmented fringes. Amongst these 

 there are from twelve to sixteen thread-like processes which are not pigmented ; each 

 terminates in a bright, highly refractive spherule, like that of a ciliary rod. These 

 spherules form a ring around the cone immediately beneath the corneal facet. They are 

 figured by Claparede *. I shall speak of these threads as " ciliary threads." 



The form and connexion of the ciliary rods and threads suggest the idea that they 

 may be concerned in regulating the amount of light which passes through the apex of 

 the cone by means of a local mechanism. The action of light on the highly refractive 

 particles possibly gives rise to a contraction of the iris. Whatever the mechanism by 

 which the size of the aperture in the iris is regulated, it is certain that it is contracted 

 in a bright light and dilated in the dusk, in nocturnal insects at least. 



In Moths the well-known luminous reflex of the eye after the insect has been kept 

 for a time in darkness is undoubtedly due to reflection from the spindles, as these are 

 surrounded, in them, by numerous very close parallel tracheal vessels, which form a 

 very perfect reflector. The disappearance of the reflex in the light is certainly due to 

 the contraction of the irides. The gradual contraction of the bright field in each lenti- 

 culus of the cornea can be observed by means of a modification of the ophthalmoscope 

 which I devised and shall hereafter describe. 



It is a remarkable fact that the irides of the peripheral segments of the dioptron close 

 before those of the central portion when the light falls directly in the axis of the latter, 

 so that when the luminous eye of a moth is observed, it appears as if the bright part 

 gradually became smaller until at last it entirely disappears. 



Kiihne is certainly mistaken when he says that the luminous reflex is of periodic 

 recurrence f , and cannot be produced in the day. After exposure to a bright light the 



L. c. 

 t W. Kiihne, " Observations on Notodon and Acherontia," Untersuch. Phys. Heidelberg, Bd. i. p. 242. 



