Whether the two cephalic ganglia be considered as really part of a single great ganglionic chain, or the reverse, they are at any rate always intimately connected with the ventral ganglia. We have already stated that the two cephalic masses are themselves closely approximated in many Insects, and may add that in some Hemiptera the first thoracic ganglion of the ventral chain is amalgamated into one body with the sub-oesophageal ganglion, {119}and further that there are a few Insects in which this latter centre is wanting. If the cephalic ganglia and ventral chain be looked on as part of one system, this may be considered as composed originally of seventeen ganglia, which number has been demonstrated in some embryos. The anatomy of the supra-oesophageal ganglion is very complex; it has been recently investigated by Viallanes[38] in the wasp (_Vespa_) and in a grasshopper (_Caloptenus italicus_). The development and complication of its inner structure and of some of its outer parts appear to be proportional with the state of advancement of the instinct or intelligence of the Insect, and Viallanes found the brain of the grasshopper to be of a more simple nature than that of the wasp. [Illustration: FIG. 65.—Brain of Worker Ant of _Formica rufa_. (After Leydig, highly magnified.) Explanation in text.] Brandt, to whom is due a large part of our knowledge of the anatomy of the nervous system in Insects, says that the supra-oesophageal ganglion varies greatly in size in various Insects, its mass being to a great extent proportional with the development of the compound eyes; hence the absolute size is not a criterion for the amount of intelligence, and we must rather look to the complication of the structure and to the development of certain parts for an index of this nature. The drone in the honey-bee has, correlatively with the superior development of its eyes, a larger brain than the worker, but the size of the hemispheres, and the development of the gyri cerebrales is superior in the latter. In other words, the mass of {120}those great lobes of the brain that are directly connected with the faceted eyes must not be taken into account in a consideration of the relation of the size and development of the brain to the intelligence of the individual. The weight of the brain in Insects is said by Lowne to vary from 1/150 to 1/2500 of the weight of the body.