138 



NA TURE 



[June i i, 1896 



where i° per x feet is the unknown gradient at the end of the 

 Glacial period. At Port Jackson, h is 32""5, and the gradient 

 x" per 80 feet. If /'be the corresponding value of/, we have 



32-5 



Jt' "■' '\x So/ 



V49 Soy 



^ - i^, ^ VC-c) 



finds K to 



4=0-56. 



Lord Kelvin, making use of Forbes' observation 

 be 400, so that the last equation reduces to 



65 _ 38 



This is satisfied if / and /' are both 2325 years, but so small a 

 length of post-Glacial time is of course inadmissible. But, if / 

 be increased beyond this value by any amount, it may be shown 

 that /' is increased by a smaller amount ; that is to say, the 

 length of post-Glacial time must be greater in the north of 

 England than at Port Jackson. 



The following table contains some numerical estimates of the 

 relative lengths of post-Glacial time in the.se districts, calculated 

 from the last equation : — 



Port Jackson. 



irth of Eiigl.-ii 



10,000 

 20,000 

 30,000 

 40,000 

 50,000 

 100,000 



4,Soo 

 6,100 

 6,900 

 7,500 

 7,900 

 9,100 



Too much stress should not of course be laid on these figures. 

 The second and third, especially, of the assumptions on which 

 they are based, must certainly be far from true. But, at any rate, 

 it seems clear that the ice nmst have left the neighbourhood of 

 Port Jackson much more recently than it left the north of 

 England. 



Whether this conclusion points to an alternation of the Glacial 

 periods in the two hemispheres, and so furnishes an argu- 

 ment in favour of CroU's theory, is perhaps doubtful. But 

 it shows, I think, how important it is, from a geological point of 

 view, that further temperature observations should be made in 

 the coal-mines and other borings of Australia, New Zealand, 

 and South Africa. C. Davison. 



PLA NT-BREEDING. 

 XXTE are most of us now-a-days so much accustomed to see our 

 gardens or our houses bedecked with flowers, and our 

 tables supplied with vegetables and fruit, that we take these 

 things for granted, and do not trouble to inquire whence 

 they come or how they are produced. But if we look 

 back even a few years, we shall see how much larger a 

 share plants have now in our lives than they had then. 

 We shall see, moreover, that while there has been enormous 

 numerical increase, there has also been in many cases continued 

 progression in form and other attributes. We are not concerned 

 here with the introductions from foreign countries, important 

 though they are ; our business for the moment lies with the 

 changes resulting from the natural processes of variation as con- 

 trolled by the art of the gardener. The garden roses of to-day, 

 for instance, are not the roses of a dozen years ago, and as to the 

 sorts that were grown by our fathers and grandfathers, they have, 

 with some few exceptions, utterly gone. It is the same with peas 

 and potatoes, and with most other plants that are grown on 

 a large scale. True, there are some exceptions ; there are some 

 "good old sorts," which seem to show by their persistence that 

 they are the fittest to survive under existing conditions. The 

 black Hambro' grape is an illustration, the old double white 

 Camellia is another ; but these plants are not reproduced 

 by seed, and therefore do not invalidate the rule, that 

 each succeeding generation of plants differs in some degree 

 from its predecessor. At first the differences are slight, 

 and it may be imperceptible to all but the trained expert ; but 

 they become more accentuated as time goes on, till at length 

 they eventuate in forms so different from that from which they 

 sprang, that they would undoubtedly be considered of specific, 

 if not of generic, rank, were not their history known. The 



NO. 1389, VOL. 54] 



Jackman Clematis and its near allies may be cited as cases in 

 point, and still more remarkable are the tuberous Begonias, 

 which, like the Clematis just mentioned, have been created, so 

 to speak, within the last quarter of a century, and which are so 

 different from anything previously known amongst Begonias, that 

 they have actually been raised to the dignity of a genus by 

 M. Fournier, a French botanist. Pansies and Auriculas — 

 garden productions both — are now, morphologically speaking, 

 as good species as are most of the groups of individuals to which 

 this rank is assigned by naturalists. Of their seedlings a large 

 proportion comes true — that is, the parental characteristics are 

 so far reproduced that there is no greater amount of variation 

 among the oflspring of many of these artificially-made species 

 than there is in the progeny of natural species. If, as is the case 

 in some Auriculas and the gold-laced Poly.anthus, we find little 

 change has occurred during the last few years, may not this 

 relative ins'ariability be the result of the gradual assumption of 

 a degree of stability which we usually associate with the idea of 

 a species? Ag.ain, it often happens that these high-bred, clo.se- 

 fertilised plants become sterile, so that their continuance can 

 only be ensured by cuttings, or some means of vegetative pro- 

 pagation. Is not this analogous to the retrogression and ultimate 

 extinction which occur in natural species ? It is not necessary here 

 to cite more illustrations ; our concern lies rather with the way in 

 which these changes are bl-ought about. This leads us to what 

 is called the improvement of plants, or plant-breeding. There 

 seems to be a growing tendency to make use of the latter term ; 

 but if it is to be adopted, it must be taken in a broad sense, and 

 not limited to the results of sexual propagation. 



The two methods, made use of Ijy gardeners and plant-raisers 

 for the improvement of plants, are selection and cross-breeding 

 — the latter, as far as results are concerned, only a modification 

 of selection. The natural capacity for variation of the plant 

 furnishes the liasis on which the breeder has to work, and this 

 capacity varies greatly in degree in different plants, so that 

 some are much more amenable and pliant than others. 

 The trial-grounds of our great seedsmen furnish object- 

 lessons of this kind on a vast scale. Very large areas are 

 devoted to the cultivation of particular sorts of cabbage, of 

 turnips, of peas, of wheat, or whatever it may be. The 

 object is two-fold — primarily to secure a "pure stock," and 

 secondarily to pick out and to perpetuate any apparently 

 desirable variation that may make itself manifest. 



The two processes are antagonistic — on the one hand, 

 every care is taken to "preserve the breed," and to neutralise 

 variation as far as possible, so that the seed may "come 

 true"; on the other hand, when the variation does occur, the 

 observation of the grower marks the change, and he either 

 rejects the plant manifesting it as a " rogue," if the change 

 is undesirable, or takes care of it for further trial, if the variation 

 holds out promise of novelty or improvement. It is remarkable 

 to note how keen the growers are to observe the slightest change 

 in the appearance of the plants, and to eliminate those which do 

 not come up to the required standard, or which are not " true." 

 Where the flowers lend themselves freely to cross-fertilisation by 

 means of insects, as is the case with the species and varieties of 

 Brassica, it is essential, in order to maintain the purity of the 

 offspring, to grow the several varieties at a very wide distance 

 apart. In passing along the rows or "quarters," the plant- 

 breeder not only eliminates the " rogues," and retains what he 

 thinks may be desirable variations, as we have said, but he 

 specially marks those plants which most conspicuously show the 

 characteristic features of the particular variety he desires to 

 increase, and he takes care to obtain seed from the plants so 

 marked. The variety thus becomes " fixed," but it is obvious 

 that that word is only used relatively ; really, there is a constant 

 change, which may be either in a retrograde direction, or which 

 may be looked on as an amelioration. Thus, in the seedsmen's 

 advertisements wesee announcements of this character : " So-and- 

 so's Improved Superlative Cucumber " or whatever it may be. 

 This " improvL-ment," when it exists, is the result of the careful 

 scrutiny, elimination, and selection exercised by the raiser. 

 These are repeated season after season, till a degree of fixity 

 is attained and a good "strain" is produced. 



Fierce competition and trade rivalry forbid the growers to 

 relax their efforts, and thus it hajipens that the pea or the 

 potato of to-day is not the same, even though it may be called by 

 the same name as its predecessors. To the untrained eye, the 

 primordial dift'erences noted are often very slight ; even the 

 botanist, unless his attention be specially directed to the matter 



