SEPTEJrBER 1, 1892.] 



KNOWLEDGE. 



173 



as many as eight or ten all touching each other, as near 

 " Eomer." These little hills, haviug oritices on the sum- 

 mits, lie directly over the cleft. The so-called 

 " river bed " does not avoid them and go round 

 the base, hence the proof that the cleft was 

 made first, and the eraterlet, or cone, arose on 

 it afterwards. There can be no doubt, there- 

 fore, that these craters and cones are piled-uj) 

 formations, and hence that their invariably 

 snow-white material has been derived fi-om 

 beneath the surface — through the cleft — at 

 isolated localized passages. If these little piled 

 cones were occasionally coloured reddish, light or dark 

 brown, or dark grey, we might suspect they were formed 

 of volcanic ejecta ; their absolutely uniform whiteness, 

 however, all over the globe, whether at the poles or equator, 

 whether on mountain regions or marea, forces us to the 

 conclusion that they are all formed of snow, exhaled as 

 aqueous vapour through orifices along the clefts, which 

 here and there pass down to warmer and moister sub- 

 strata. Thus, while this frequent structural association of 

 craterlets and crater rows, with clefts, is intelligible, if we 

 look on the latter as profound fissures in the crust, it is 

 latterly imintelligible if they are viewed as water-courses. 



On the moon, as on our earth, there is yet evidently a 

 vast store of internal heat ; as this is slowly dissipated, 

 and induces internal contraction, the lunar marea, like our 

 tcrrc.-itrial sea hottoiiis, slowly sink in, fracturing the outer 

 crust, as we have seen, by doing so, in several different 

 ways, as in A, B, and C ; showing us also that the outer 

 " crust " of the globe is of a somewhat brittle nature and 

 of considerable thickness ; and, lastly, that the dissipation 

 of hoat-energy is largely effected all over the surface, by 

 the exudation of aqueous vapour through pores or passages, 

 over and around which it piles as minute cones or 

 " craterlets " of snow. 



From the equator to the poles they are seen in count- 

 less thousands as a late or recent feature, and one of 

 them quite naturally arising in Linue would explain the 

 change seen therein, and solve that celebrated mystery. 

 Sibsagar, Assam. S. E. Peal. 



[The evidence brought forward by Mr. Peal with regard 

 to the general subsidence of the great lunar marea seems 

 to me conclusive. The passage from Mr. Nelson's book 

 on the Moon, quoted in the June number of Knowledge 

 (p. 115), does not refer to the rills or clefts as river beds, 

 but only speaks of them as bearing " some resemblance " 

 to water-courses, and as frequently commencing " at the 

 end of a system of branched valleys leading from a high- 

 land." Mr. Neison says (p. 72), " With regard to the true 

 nature of these rills or clefts absolutely nothing is known, 

 whilst they are too delicate objects to allow much, if any, 

 of the detail of their formation to be made out. It has 

 been supposed they are cracks or fractures in the lunar 

 surface ; but their intersection and general conditions of 

 existence seem quite inconsistent with such a supposition, 

 more especially in their behaviour with reference to the 

 various formations they pass through, round, or over. In 

 many points they bear some resemblance to the dried beds 

 of lunar water-courses or rivers, but in many features do 

 not seem in accord with such an origin, though perhaps it 

 presents the most feasible explanation of their nature of 

 all ; but their true nature will not be ascertained until they 

 have been made the subject of a searching examination 

 with a powerful telescope of the highest excellence, and 

 thus details of the method of their construction have been 

 obtained. Perhaps, unlike the terrestrial river beds, these 

 rills may have arisen independently, but have served after- 

 wards the purpose of river beds ; their connection with the 



system of delicate valleys renders such a view somewhat 

 probable. Thus many of these rills commence at the end 

 of a system of branched valleys leading from a highland, 

 whilst others can be detected winding along the bottom of 

 extensive valley regions. At other times they appear 

 however, entirely independent of such formations."] 



THE OLDEST FISHES AND THEIR FINS. 



By R. Lydekker, B.A.Cantab. 



IN the article on " Mail-clad Animals," published some 

 time ago in Knowledge, it was shown how the fishes 

 of the more ancient periods of the earth's history 

 were frequently characterized by having their bodies 

 protected by a coat of armour, and also how this 

 armour has been lost by most of their modern descendants. 

 At the same time it was mentioned that a few of these mail- 

 clad fishes, like the gar-pike of the rivers of North America, 

 and the many-tinned bichir (roli/jiterus) (Fig. 1) of the 



Fio. 1.— The Bichir. 



Upper Nile, still linger on, as if for the purpose of showing 

 us what their ancestors were like. 



In addition, however, to their armour, and its gradual 

 loss with the advance of time, there are many other points 

 of view from which these ancient fishes are of more than 

 ordinary interest, and we accordingly propose in this 

 article to consider the curious modifications which have 

 taken place in the structure of their fins as we ascend in 

 the geological scale. We shall, moreover, be led to notice 

 briefly one of the most remarkable types of fossil fish 

 teeth found in the older secondary rocks, since it proves 

 that one of our living fishes is the oldest kind of 

 vertebrate now inhabiting the earth. 



Before going further, we must mention that existing 

 fishes have been divided into several main groups, dis- 

 tinguished from one another by structural peculiarities. 

 One such group includes the sharks and rays, charac- 

 terized by their cell-like gills and scaleless bodies. Then 

 we have the smaller group of lung-fishes, now represented 

 by the baramunda, of Queensland (figured in the article 

 on " Mail-clad Animals "), and the mud-fishes of the rivers 

 of Africa and South America, all of which can breathe 

 either by gills or by lungs. Another group is formed by 

 the so-called ganoid fishes (Fig. 1), many of which have 

 the bony armour already mentioned ; while the great 

 majority of the fishes of the present day, although nearly- 

 related to these ancient ganoids, have been generally 

 separated as a distinct group, under the title of bony 

 fishes. That name they take from the circumstance that 

 their skeletons are fully ossified, and do not partake of the 

 cartilaginous nature of those of a shark or a ganoid. 



Now if we look at the paired fins (or those which 

 correspond with our own limbs) of any ordmary bony fish, 

 such as the perch (Fig. 2), we shall see that they are 

 formed of a number of bony rays, starting from a single 

 point of origin, and thence spreading out in a fan-like 

 manner. We shall also not fail to observe that the tail of 

 such a fish has a very similar kind of structure, likewise 

 consisting of bony rays, symmetrically arranged, and 

 starting from a curved line where the scales suddenly stop. 

 We may also see from the figure of the skeleton of such a 

 fish that the backbone likewise stops suddenly where the 



