July 28, 1893.J 



SCIENCE. 



45 



the other (pendent) longer, thinner, straight, and appressed closely 

 downwards to the stem ; the leaves on the branches being closely 

 imbricated all round. The stem bears leaves very different in form 

 and structure from those of the branches. 



Now Sphagnum acidifolium is a most variable moss; the list 

 of recognized species in Europe alone numbering about thirty. 



Among these are several distinct and well-marked forms, such as 

 the following : In one the branch leaves, instead of being straight and 

 closely imbricated as described above, are bent back in the middle 

 and spread almost at right-angles from the branch — the forma 

 squarrosa. In a second the branches, instead of being straight or 

 nearly so, are hooked or contorted — the falcate variety. In a 

 third, the forma compacta, the whole plant takes a short, com- 

 pact habit, the stems being much shortened and closely tufted, 

 the fascicles of branches close together, and the branches them- 

 selves short and stunted, with the leaves closely set. In a fourth 

 the differentiation between the stem and branch leaves almost or 

 quite disappears, the former acquiring the form and structure of 

 the latter, the forma homophylla, and so on with two or three 

 more distinct varieties. 



Now, if we turn to the other species of the genus, we find that 

 of those found in Europe and North America there is hardly one 

 which does not include one or more of these six or seven distinct 

 varieties which we find in S. acutifolium. Thus of nineteen 

 European species (all but two of which are natives of North 

 America; sixteen, and perhaps eighteen, have varieties belonging 

 to the foi'ma compacta, fourteen at least, and perhaps four others, 

 have the squarrose variety, and so on to a greater or less degree 

 with the other forms. At least two of these forms are found under 

 every one of the species, and in more than one species all the 

 forms are found. 



Here we have a clear case of analogous variations. It cannot 

 be supposed that they are instances of reversion to a common an- 

 cestral form, for, apart from other considerations, the variation 

 in some of the forms Is in a directly opposite direction to that 

 which it takes in others. The delicate, elongated forms of the 

 teiiellm and the dense, compact forms of the compactce can hardly 

 both be reversions to a common ancestral type ! 



So far we have exactly the same thing that we see in many 

 races of domesticated species, such as Darwin has pointed out, for 

 instance, in the races of the domestic pigeon; but we do not often 

 see it carried out on such a wide and instructive scale. 



But what is of especial interest in the case of the Sphagnaceee 

 is that, when we go further and consider the characters that dis- 

 tinguish the different species from one another, we find that the 

 very points which we have seen mark off the above varieties (and 

 render them, as a rule, more distinct than the other varieties of . 

 the species) are in several cases those which are most characteristic 

 in separating from one another the species themselves. Thus S. 

 squari'os^im is specially marked by the spreading leaves ; S. rigidum 

 has for its most obvious features the very characteristics by which 

 the compacta forms above described are distinguished ; S. siibse- 

 eundiim in most of its forms is marked by its falcate or contorted 

 branches; while a group of species, classed by Lindberg as 

 Homophylla, are characterized by that similarity of stem and 

 branch leaves which I have described above as the feature of the 

 corresponding variety; and so on with the other forms. Here we 

 have exactly fulfilled the supposition of Darwin quoted above, 

 "that a variety of one species would resemble in certain charac- 

 ters another and a distinct species," and fulfilled, too, on a scale 

 which, at any rate, precludes the possibility of its being due to 

 fortuitous coincidence. 



On any theory of creation that did not presuppose a common 

 ancestry for these species of Sphagnum, it might indeed be possi- 

 ble to account for the analogy between the varieties of different 

 species by assuming the variations to be the direct results of the 

 environment (a more than doubtful assumption, moreover-); but 

 the more we lay this cause under contribution to account for the 

 varietal forms, the harder it is to believe that precisely the same 

 variations in the species, only cai-ried out to a higher degree of 

 permanency, are due to entirely different and quite unconnected 

 causes. 



The above facts appear to me to form a peculiarly interesting 



support to Darwin's argument from analogous variation. In the 

 first place, the possibility of reversion is, as I have pointed out, 

 eliminated, and reversion and analogous variation, which are 

 quite distinct principles, are too often indistinguishable in their 

 results for us to be quite certain that we have a genuine example 

 of the latter. In the next place, as Darwin points out, analogous 

 variations are liable to be eliminated as not being necessarily 

 serviceable; that they are not eliminated in the Sphagna is, I be- 

 lieve, partly due to the peculiar conditions under which these 

 plants usually grow, but this opens too wide a field to enter upon 

 here. In addition to these reasons, we have here an Illustration 

 drawn from species and varieties in a state of nature; examples of 

 analogous variations have usually to be drawn from domesticated 

 forms, where their value is limited by their necessarily applying 

 to races and varieties only, and not to distinct species. 



I append a table (taken from Jensen's paper quoted above), which 

 shows at a glance the distribution of these varietal forms among 

 the European species of Sphagnum. A f indicates the existence 

 of the variety heading the column under the species opposite to 

 which it is placed ; a ? means that the existence of such a form is 

 probable, but is insufficiently attested or not clearly enough 

 marked to be entered as certain. It must be remembered that 

 there is always a possibility of gaps being filled up by future re- 

 search, but the table is, I think, as it stands, sufficiently striking. 



Group. 



Spliagnum laxlfoUum, C. M 



" intermedium, Hoffm. 



** riparium, AngBtr 



" lindbergll, Schlmp 



" wulfii, Qirg 



" acutifolium, Ehrb — 



" strlctum, Lindb 



" flmbriatum, Wils 



" teres, Acgstr 



" squarrosum, Pers 



" subsecundum, Nees.. 



*' caricinum. Spruce 



" tenellum, Bhrb 



" compactum, D. C 



" molle, Sull 



" aDgstromll, C. Hartm 



** cymbifollum, Ehrb. .. 



" papillosum, Lindb 



" austinl, Sull 



ogjo"a-j aaoa 



THE CLOSE OF THE ICE AGE IN NORTH AMERICA. 



BY R. W. MCFABLAND, LL.D , LATE PRESIDENT OF MIAMI tTNIVEBSITY. 



This is a question of interest to scientific men in general, and 

 to geologists and glacialists in particular. 



In Professor Wright's "Ice Age in North America," p. 448, in 

 speaking of Croll's table of the eccentricity of the earth's orbit, he 

 says : " According to this table the modern period most favorable 

 to the production of a glacial epoch began about 240,000 years 

 ago, and ended 70,000 years ago." Again, on p. 450, we have 

 this: " If, therefore, the glacial period should prove to have ended 

 only 10,000 years ago, instead of 70,000, the Darwinian would be 

 relieved of no small embarrassment." 



A genuine scientist, of course, has no preconceived theory to 



