296 MEMOIES OF THE NATION^AL ACADEMY OF SCIENCES. 



He finds in the protocerebrvm of the insect (locust, etc.) all the constituent parts of the 

 protocerebrum of Crustacea. These are the layer of postretinal fibers, the ganglionic plate, the 

 external chiasma, the external medullary mass, the protocerebral lobe, the ocellar nerves and 

 lobes, the bridge of the cerebral lobes, and the middle (central) protocerebrum. 



The deutoeerehruni comprises the olfactory antennal lobes, and the dorsal lobe of the deuto- 

 cerebrum. The deutocerebrum gives origin to four pairs of nerves, i. e., (1) the antennary nerves; 

 (2) the accessory antennal nerves; (3) the tegumental nerves; (4) the root of the stomatogastric 

 ganglion. 



The tritocerehrxim is formed of a pair of lobes (tritocerebral lobes), which are united by a trans- 

 verse commissure passing under the (^esophagus. The lobes give origin to a nervous trunk, 

 which is divided into two branches, i. e., the labral nerves, and root of the frontal ganglion. 



Dr. Saint Remy adopts Viallanes' classification, but he concludes that in Myriopoda the 

 bi"ain is divided into the "three ganglions which are ])erfectly comparable to the brain of the 

 higher Crustacea and of insects, though their structure is infinitely more simple." But in the 

 brain of Arachnida he finds only the protocerebrum, and adds: '■'■ Mais pour ce qui est de la seconde 

 region du cerveau, ou ganglion rostromandibulaire, la determination des homologies est plus delicate.''' 

 The deutocerebrum of the insects, myriopods and crustaceans " n''est pas represents id.''' He thinhs 

 that the rostro-mandibular ganglion of Arachnids is the homologue of the third region or trito- 

 cerebrum. 



The shape and relations of the brain of insects is well shown by Fig. V. 



Coming to Limulus we shall see that the brain does not include the rostromandibular ganglion 

 found in spiders and scorpions, and that neither the tritocerebrum or the deutocerebrum is found, 

 the brain representing the protocerebrum alone ; or if the deutocerebrum is present it is not pre- 

 cesophageal and is not yet fused with the archicerebrum.* 



IV. Histology. 



The histological elements of the brain of Limulus are closely similar to those of the Arachni- 

 dan brain, so ably described by Saint Remy. As in other Arthropods the histological constituents 

 of the brain are the large ganglionic cells with abundant protoplasm, he punctuated or myeloid 

 substance, the nervous fibers which form the origins or roots of the nerves, and the small gangli- 

 onic cells or nuclei. 



These last-named cells unite to form certain bodies, the real nature and homology of which 

 has made the study of this subject especially difficult. Indeed, the most striking feature of the 

 brain of Limulus are what we called in our former memoir the convoluted, ruffle-like "nucleoge- 

 nous bodies." They form cortical layers enveloping the cerebral and optic lobes, and where scat- 

 tered inclose masses of fine fibrillar and myeloid or punctuated substance. These bodies appear 

 at first sight to be composed of closely crowded nuclei, but when they are scattered they are seen 

 to be small ganglionic cells. They appear to be the "ganglionic nuclei" {gangUiisen Kerne) of 

 Dietlt (p. 303, etc.) which he also refers to as "free protoplasmaless nuclei" (p. 496). See also 

 Dietl's later paper (1878), where he thus defines these ganglion cells: "4. Ivleine Zellen mit sehr 

 schmalen Protoplasmasaum der triibe Nucleus enthiilt Nucleoli (Sinnesanschwellungen)," p. 14. 



I also regard them as identical with the '■'■cellules chromatiques'' of Saint Remy. He describes 

 them as "unipolar cells, so poor in protoplasm that their cellular outlines escape observa- 

 tion in sections," and adds: "Besides the extreme reduction of the protoplasmic body these 

 elements present a remarkable peculiarity : the richness of their nucleus in chromatic substance 

 capable of taking very distinctly the coloring matters, even after the prolonged action of osinic 



* The terms protocerebrum, deutocerebrum, and tritocerebrum are rather cumbersome, and as they represent 

 three distinct pre-tesophageal segments (arthromeres), it seems to us more convenient to call the i>rotocerebrum the 

 optic segment or division of the brain, because the principal lobes composing it are those sending nerves to the com- 

 pound eyes and ocelli; so also the deutocerebrum might be denominated the antennal segment, and the tritocere- 

 brum the labral segment. This nomenclature would well accord with the names mandibular, and tirst and second 

 maxillary segments situated behind the mouth, the ganglia contained in them being called by the same names, man- 

 dibular, etc. 



t Die Oiganization des Arthropodengehirns. 



