Abstract
In the present paper the writer gave a detailed analysis and comparison of the bones, especially of the cranium and some of the visceral skeletons in 5 species of the Japanese mugilid-fishes referred to three genera, Mugil, Liza and Chelon. By this study it was found that the osteological characters provide more clear specific and generic criteria and also more reliable indication of phyletic relation than the older schemes which based essentially on the external features.
In determining the relationship of the genera one another, the following characteristics are especially taken into consideration ; 1) presence or absence, and degree of differentiation, if present, of the basisphenoid;2) shape and degree of developement of parasphenoid, especially in the middle and posterior portions; 3) with or without the posterior opening of myodome; 4) presence or absence, and shape, if present, of the basioccipital process; 5) presence or adsence of elongated posterior-zygapophysis of the second vertebra; 6) form of the urohyal; 7) configulation of the post-temporal; 8) form of ethmoid region: and 9) form of preorbital.
The diversities of these organs have been simply divided into two categor-iies, viz, plus and minus, in each element, (Table 1 and 2). As indicated in these tables, the genus Chelon is, as a whole, most undifferentiated representative of the mugilids, although this fish having the remarkable basioccipital processes and peculiar shape of premaxillary. The genus Mugil is unquestionably more differentiated than the others in lacking both the basi-sphenoid and the posterior opening of the myodome, in having well developed parasphenoid and ethmoid, a pair of long posterior projection of urohyal, and three remarkable articular processes on post-temporal.
The genus Liza also more specialized than the genus Chelon at least in having peculiarly elongated posterior-zygapophysis in second vertebra, well developed ethmoid, three articular process on post-temporal, and in lacking the posterior opening in myodome, although premaxillary less developed than in the latter.
It is interesting to note that the genus Chelon, despite mostly generalized in the mugilid-fishes, provided with a few remarkably specialized characters as aforementioned. Is it not very reasonable, therefore, to think that the genus Chelon is rooted very deep in phylogeny within this group?
