336 Dr. C. Liitken on the Limits and 



bility tliat future discoveries may some day demonstrate to us 

 these still unknown bonds*; nor must we forget that it is not 

 many years since naturalists did not hesitate to refuse a place 

 among the Ganoids to \\\q Asjndorhynchij the Godacanthi^ and 

 the Pycnodontes^ which we now arrange without hesitation 

 among the undoubted Ganoids. 



a. In the first place, the Lepidosirens or Proto2)teri, classed 

 by some writers of incontestable authority with the Ganoids, 

 but most frequently regarded as forming a peculiar subclass 

 [Dipnoi)^ will form, in my opinion, only an aberrant tribe or 

 a suborder of the Physostome Teleosteans, to be placed in the 

 immediate vicinity of the Ganoids and particularly of tlie 

 Crossopteri {P/ianero2)Ieuro7ij for example). 



b. Then the Sturgeons are also Physostome Teleosteans, 

 which should be arranged as near as possible to the Chon- 

 drosteij between the latter and the Gano'ideij with which, 

 however, they must not be united f. 



c. The Amice approach the Ganoids and Chondrosteans by 

 a number of remarkable anatomical peculiarities ; but Ave 

 should not be more justified in classing ^?;wa with the Ganoids 

 than in arranging the Sturgeons among the Selachia. It is a 

 special type, belonging to the true Physostome Teleosteans, 

 leading towards the Ganoids, but not attaching itself to them. 

 Moreover the removal of this group from the suborder Ga- 

 noidei will but slightly modify the palasichtliyological system, 

 as it includes only a small number of forms [Notceus^ 

 Cycluriis^ Aviiojtsis) , which perhaps ought to be united with 

 Amia itself. 



d. There is also no positive reason for arranging the Jurassic 

 Teleostei [Lejitolejyidesj Megalari^ and Caturi) either with 

 the Amiidas or with the Ganoidei. If we consult the synthetic 

 •method, it will lead us rather towards the Haleco'ides — that is 

 to say, the Salmons, Heri'ings, and Clupesoces. They are 

 consequently true Physostome Teleosteans, and, with the ex- 

 ception of the Belonorliynclnis &c. of the Trias, the most 

 ancient representatives of this suborder. Moreover it will be 

 impossible to separate the three families above named from 

 each other ; those who, with the modern palieichthyologists, 

 Heckel, Wagner, and Pictet, place the Lepytolejpides among the 

 true Teleostei, will be obliged likewise to place there the 

 Megaluri and Caturiy notwithstanding the fulcral scales bor- 

 dering their fins ; the filiation of the species, the crossing of 



* At this moment the journals inform us of the discoveiy in Australia 

 of a new genus of freshwater fish, intermediate between the Lepidosirens 

 and the Palaeozoic Dipteri ! [See papers by Dr. Giinther and Messrs. Han- 

 cock & Atthey in the March Number of this Journal. — Ed. Ann. Nat. Hist.'] 



t The affinities of the fossil genus Chondrostcus are perhaps still 

 doubtful. 



