May 5, 1892] 



NA TURE 



II 



mass of plants would be so great that the new leaves 

 budded out would find no room in which to expand ; but, 

 by virtue of one very simple provision, viz. the existence 

 of inequalities of level along the edges of the leaves, 

 clear spaces and lanes are left between the floating 

 leaves, so long as any unoccupied space remains. 



Long exposure to the air, especially in still weather, 

 affects the life of duckweed in a material way. Dust and 

 decaying organic substances give rise to a pellicle, which 

 is most mischievous to floating plants : and I think I could 

 show, if time allowed, how much the habits of duckweed 

 have been altered thereby. But, apart from visible im- 

 purities, mere exposure to air gives, as Lord Rayleigh 

 has taught us, a considerable degree of superficial vis- 

 cosity to water. Hence, the leaves of duckweed, when 

 the surface is contaminated, will tend to lie in whatever 

 positions they may be thrown by accidental causes, such 

 as wind, and the attractions due to capillarity will be 

 more or less impeded. But the effect of the superficial 

 viscosity will in time be overcome by the attractive forces, 

 so that it probably does not in the long run greatly affect 

 the distribution of the leaves over the surface of water. 



Many other floating plants, but not all, behave more or 

 less like duckweed, and for the same reason. As yet I 

 know of none which space themselves quite so effec- 

 tually, and the extreme abundance of the common duck- 

 weed, as well as its world-wide distribution, may be partly 

 due to the completeness of its adaptation to capillary 

 forces. Some dead objects may accidentally take a shape 

 which causes them to spread out over water, but I have 

 met with none which have particularly struck me. Float- 

 ing natural objects, such as sticks or seeds, behave, in 

 many cases at least, very differently, and become densely 

 massed. My attention was first called to this subject by 

 seeing how different was the grouping of duckweed from 

 that of some seeds of Potamogeton naians, which were 

 floating in the same pond. 



The capillary forces which spread the leaves of duck- 

 weed or Azolla upon the surface of water are indirectly 

 concerned in the transport of these and like plants to 

 fresh sites. If we put a stick into water overspread with 

 duckweed, we cannot fail to notice how the leaves cling to 

 the stick. They cling in a particular way, which enables 

 them to bear transport more safely. The wetted surface, 

 for obvious physical reasons, is attracted to the wetted 

 stick ; and the water-repellent surface, which is that which 

 best resists drying, is outwards. The tenacity with which 

 duckweed clings to the legs of water-birds, and the 

 position which it almost inevitably takes under such cir- 

 cumstances, may have a good deal to do with the safe 

 transport of the plant to distant pools. It is not, I think, 

 too much to say that the prosperity of duckweed depends 

 very largely upon the capillary forces which come into 

 play at the surface of water. 



We have now exhausted our time, though I have been 

 obliged to leave unnoticed many special adaptations of 

 living things to the peculiar conditions which obtain 

 on the surface of water. Had time allowed, I should 

 have been glad to say something about the aquatic 

 animals which creep on the surface-film as on a ceiling, 

 and about the insects which run and even leap upon 

 the surface-film without wetting their minute and hairy 

 bodies.i All small animals and plants which float on 

 water necessarily come into contact with the surface-film, 

 and have to deal with the difficulties which result from 

 it. We have seen that they generally manage in the long 

 run to convert these natural difficulties into positive 

 advantages. 



1 have to thank my colleague, Dr. Stroud, for his fre- 

 quent explanations of the physical principles upon which 

 ttiese adaptations depend, and also for much practical 

 and valuable help in the preparation of suitable experi- 

 ments. 



' See Nature, vol. xliv. p. 457. 

 NO. I 175, VOL. 46] 



THE DISCOVERY OF AUSTRALIAN-LIKE 

 MAMMALS IN SOUTH AMERICA. 



'"PHE year 1891 proved a notable one in regard to 

 *■ marsupials. The existing mole-like marsupial 

 {Notoryctes^ from the deserts of Central Australia 

 having been made known to us, news came of the dis- 

 covery in the Tertiaries of Patagonia of remains of car- 

 nivorous marsupials closely allied to the existing pouched 

 wolf, or Thylacine, of Tasmania. This discovery was 

 immediately recognized as one likely 10 considerably 

 modify some of our views regarding the distribution of 

 mammals. A preliminary account of these new marsu- 

 pials was given by Dr. Florentino Ameghino in a paper 

 written for the new serial, Revist. Argent. Hist. Nat. 

 This description seems to leave no doubt as to the cor- 

 rectness of the diagnosis of the fossil remains. 



Before going further, it may be well to remind our 

 readers that, with the single exception of the opossums 

 {DtdelphyidcE) of America, all marsupials are now ex- 

 clusively Australasian The carnivorous types, such 

 as the Thylacine {Thylacinus) and the Tasmanian 

 Devil {Sarcophilus), are distinguished from all living 

 mammals in that their upper cutting-teeth (incisors) 

 are either four or five in number on either side, while 

 in the lower jaw there are invariably three. This rela- 

 tion is shown in the figure of the skull of the Ta5- 



Front view of the skull of the Tasmanian Devil. (After Fiower. ) 



manian Devil— a near ally of the Thylacine— where, 

 between the large tusks of the upper jaw, we see the 

 four pairs of incisors opposed to only three pairs in 

 the lower jaw. In ordinary mammals, on the other 

 hand, the number of pairs of incisors in each jaw does 

 not exceed three, the number of those in the two jaws 

 being usually equal. A further peculiarity of marsupials 

 is that the cheek or grinding teeth comprise four true 

 molars and not more than three premolars ; whereas in 

 ordinary mammals the typical number is three molars 

 and four premolars, there being no known instance of 

 the presence of four true molars except in some indi- 

 viduals of the fox like Otocyon. Another peculiarity of 

 most marsupials is the distinct inflection of the lower 

 posterior extremity, or " angle," of the lower jaw, while 

 very frequently the bony palate of the skull has unos- 

 sified spaces. 



The new forms described by Dr. Ameghino were ob- 

 tained from the lower part of that great series of freshwater 

 formations with which so large an area of South America is 

 covered. It has been inferred that the Patagonian deposits 

 in question are as old as the Lower Eocene of Europe ; 

 but, although they are undoubtedly of considerable age, 

 this inference can scarcely be regarded as an established 



