80 MisceUa?ieoifs. 



to enter the parenchymatous layer of the retinal segment through a 

 connective boundary-layer filled with large nuclei ; another portion 

 of nerve-fibres originates, however, from the coating of ganglion- 

 cells itself, and crosses the first set of fibre-trains in an oblique 

 direction. Tn comparison with the e3'e-ganglion of the Malacostraca, 

 the ganglionic cortex and intercrossing of fibres are very simple, 

 and the parenchymatous mass is not yet, as in them, divided into 

 two or three parenchymatous layers between which the fibre-trains 

 form new internal crossings. 



The fibre-crossing in the eye of Branchipus therefore only repre- 

 sents the crossing distinguished by Berger as " external " in the eye 

 of the higher Crustacea. 



This considerable simplification, which we cannot assume to be 

 due to any secondary reduction, justifies us in starting from the 

 Phyllopod-eye in estimating the two main divisions of the ganglionic 

 apparatus. The first, or proximal part, which, in the eye of the 

 higher Arthropoda undergoes a further division, is the cerebral 

 portion of the eye-ganglion ; the distal portion, which is bent almost 

 at right angles to this, and which retains essentially the same 

 structure throughout, is its retinal part, or the retinal ganglion. 



This interpretation, already set up by Berger, which at the same 

 time recognizes in the ganglion-cell coat of the proximal eye- 

 ganglion a centre of projection of the second order, is perfectly in 

 accordance with the simplified structural conditions of the eye of 

 BrancMpus, in opposition to the interpretation of other naturalists 

 who, in the compound eye of the Decapoda and Insecta, do not sepa- 

 rate the retinal ganglion distinctly from the eye-ganglion, and regard 

 it as equivalent to the preceding division, but either treat the whole 

 as the retina, or, going to the ojjposite extreme, refer it to the cere- 

 brum, and regard the nerve-bundles passing to the rods only as the 

 visual nerve-fibres. 



The structure of the eye in Branchipus is also simjjler than in 

 any other pedunculate eye. Above all we have to notice the 

 absence of special pigment cells in the vicinity of the nerve- 

 rods, as also of the crystalline cones. The pigment is deposited 

 rather in the deeper hypodermal cells employed as parts of the 

 sensitive apparatus, in the elements of the nerve-rods around 

 the rhabdoma, and peripherally in the nerve-fibres of the so-called 

 nerve-bundle layer. The rapid movement of the blood takes place 

 in the interstices of the latter, and also in front of the basilar layer 

 in spaces between the attenuated ends of the crystalline bodies. 

 There is no facettation of the cornea, but, as in the eye of Phro- 

 nima (and this is the case also in that of Apus), there is a special 

 layer of hypodermal cells above the crystalline bodies. We shall 

 have to regard the presence of this layer of cells as well as the 

 absence of corneal facets and special pigment-cells, and the presence 

 of interstices for the circulation of the blood in the nerve-bundle 

 layer and the layer of crystalline cones, as representing the original 

 form of the Arthropod compound eye, and the appearance of corneal 

 facets by the deficiency of the superficial hypodermal layer as secon- 

 dary. — Anzeiger der h. ATcad. Wiss. Wien, March 18, 1886, p. 60. 



