August 22, 1895] 



NA TURE 



599 



OUR ASTRONOMICAL COLUMN. 



TflE C'KI.OSTAT. — The name calostat has been given by 

 M. (i. Lippmann to a modified fonn of siderostat which he has 

 devised (Comples reiidus. No. 19, 1895, and Observatory, 

 August). The special feature of the instrument is that it gets 

 rid of the rotation of the field of view which disqualifies the 

 siderostat for some purposes, such, for instance, as long-exposure 

 ])hotogra])hy. It consists simply of a mirror with its plane 

 ]>arallel to the earth's axis, and turning on a polar axis once in 

 forty-eight hours in the same direction as the apparent diurnal 

 motion of the heavens. It is easily demonstrated that the image 

 of any star whatever will be seen stationary in a mirror so 

 mounted, and a telescope pointed at the mirror in any direction 

 will have a constant field of view. The telescope being directed 

 to the crelostat in a given position, to observe other objects 

 having the same declination as that in view, it will only be 

 necessary to turn the mirror ; but for objects with different 

 declinations the telescope must also be moved. If it be desired 

 to use a horizontal telescope, it must be directed to the point on 

 the horizon where the object rises, and the mirror must be 

 started in a position suited to the hour-angle ; but there is a 

 limit to the use of a horizontal telescope. It is pointed out that 

 the simplicity of the instrument makes it possible to turn it into 

 one of great precision ; stability being readily attained, while the 

 l^ossibility of flexure can be reduced to a minimum. 



Adams' Masses of Jupiter's Satellites. — A question 

 having been recently raised by Mr. Marth as to the work of 

 .\dams on Jupiter's satellites. Prof. K. A. .Sampson has stated 

 the results of an inspection of the MSS. with reference to this 

 subject (Observatory, August). It appears that when engaged 

 upon a revision of Damoiseau's tables in 1875, with a view to 

 their continuation, Prof. Adams determined the following 

 revised values for the masses of the satellites : — 



in = 0-0000283 1 13 



in' — 0'00O0232355 



m" = O'oooo8i2453 



m'" = 0'oooo2i48So 



" There is no reason to suppose that Adams attached any weight 

 to the above determinations of the masses, seeing that he never 

 published the values directly ; the MS. appears to be little more 

 than a study such as he was in the habit of making upon any 

 work that he was examining, in order to test by cross verifica- 

 tions the accuracy and consistency of the whole. . . . Con- 

 siderable expectations have been built upon the fact that Adams 

 was engagecl more or less closely for some years upon the theory 

 of Jupiter's satellites. It will be well to say at once that the 

 chief fruit of his attention was published in the Nautieal Almanac 

 of 1S80 ; this, like all the rest of his published work, v.as the 

 result of e.xhaustive labour, quite out of relation to the unpre- 

 tentious form in which the outcome was presented, and only 

 ■discoverable by searching tests." 



ATMOsrnERir Refraction. — The ordinary application of 

 Vessel's expression for refraction requires that five i|uantities be 

 taken from specially prepared tables, but Prof E. C. Conistock, 

 Director of the Washlnirn Observatory, has worked out a simple 

 fornuila for conqiuting the refraction -vithout the aid of tables. 

 A transformation of Bessel's formula, and the introduction of 

 numerical constants from the Pulkowa refraction tables, leads to 

 the following simplified form : 



R = [2 -992 1 5] tanZ 



455 '9 + / 



logK = - (42-3 -f o"l2/)tan'-Z. 



The number in brackets is a logarithm ; B is the barometric 

 pressure in Knglish inches reduced to freezing-point : / is the 

 temperature in clegrees I-'ahrenheit, and Z is the zenith distance 

 for which the refraction is required. The formula for F gives 

 the logarithm in units of the fifth decimal jilace. 



The comi)utation by the formula is not more laborious than 

 the direct use of the tallies, and a comparison of the two methods 

 shows that the dift'erences in the results are far less than the un- 

 certainty in the tabular numbers themselves. Prof. Comstock's 

 pajier forms one of a series of interesting " Studies in .Spherical 

 and Practical Astronomy," in the Bullcliii of the University of 

 ■Wisconsin (vol. i. No. 3). 



NO. 1347, VOL. 52] 



ON THE ORIGIN OF EUROPEAN AND 

 NORTH AMERICAN ANTS. 



QUESTIONS belonging to zoogeography may be practical or 

 theoretical, actual or genetic ; ultimately the resolution of 

 them, whatever they may be, lakes its chief interest from their 

 relations to genetical problems, that is, to the explanation of 

 the origin of actual fauna-, and to the knowledge of the original 

 home of phyletic groups, and of the ways followed in their 

 gradual diffusion over the whole or part of the world. To this 

 purpose, not only living animals, but also fossils, have to be 

 determined, and their affinities exactly worked out ; changes in 

 the distribution of land and sea and in the shape of continental 

 areas must be investigated, and analogies and differences in the 

 difi'usion of various groups of living beings taken in considera- 

 tion, as far as they are known. The work involved is long and 

 difficult, and its results will form the science of the future. 



In a jiaper published! in 1S91, t»n the fossil ants of .Sicilian 

 amber,' I made out that at the beginning of the Miocene epoch, 

 North and South Europe had very different fauna; of ants, the 

 Sicilian amber containing genera which belong to the actual 

 Indian and Australian fauna, but wanting the typical holarctic 

 genera Formica, Lasiits, Alyrniica, which are found in the 

 Baltic amber, some species of them being extremely common 

 and abundant. A similar, but not such a striking, difference exists 

 between recent Mediterranean and North European ants, the 

 former including a greater percentage of Indian and cosmopolite 

 forms, and an absolutely and relatively lesser number of typically 

 holarctic ones, the most species of Formica, Myrmica, and 

 Lasitis not having reached .\frica (F. fiisca, L. , and M. scabri- 

 nodis, Nyl., are introduced in gardens in Algeria), and these 

 genera being scarcely represented in Mediterranean islands. 

 After discussing these facts, I came to the conclusion that South 

 Europe should have had in the Tertiary epoch an ant fauna 

 compound of old Mesozoic cosmopolite genera (chiefly Ponerinte), 

 mixed with Indian-Australian forms. In North Europe these 

 lived together with northern genera, which, after the emergence 

 of the bottom of the middle European sea, invaded the South, 

 being perhaps expelled from the North by gradual cooling of 

 climate. Later, the glacial epoch destroyed in Europe nearly 

 all the rest of tropical insects, their return being made im- 

 possible by the natural barriers of sea, deserts, and mountains, 

 accumulated southward and eastward of our continent. 



These studies I have carried a step further in a revision, now 

 printed, of the Foniiicid.-e of North America." A great number 

 of North American ants are specifically identical to European 

 ones. My attention was directed to find differences between 

 American and European specimens, and indeed but a few 

 species were so similar to their European relatives as to be not 

 distinguishable as sub-species or varieties. The one genus, 

 Epcecus and two sub-genera are exclusively Nearctic ; all the 

 other genera of North American ants not represented in Eurasia 

 (Discothyrea has two species only, one in North America, 

 another in New Zealand)are Neotropical. The northern regions 

 of Europe has the one peculiar genus Attergates, allied to 

 Epacus : middle and south Europe have two further genera not 

 foimd in other parts of the world, and some others known from . 

 the Indian region. All these facts lead to the result, that the 

 Pala;arctic ant-iauna is made of cosmopolite -f Arctic -h Indian 

 elements ; that the Nearctic fauna is similarly composed of 

 cosmopolite -f Arctic -t- Neotropical ones. 



The question that now arises is : how has such a mixture been 

 efi'ectuated — what changes have determined it ? A complete and 

 detailed answer I believe to be at present impossible ; but the 

 knowledge of the fossil mammals may help us greatly, supply- 

 ing for the want of evidence taken from fossil ants, other than 

 the .Miocene fauna of European amber, the fossil prints of 

 Kormicidce being too imperfectly known, and a careful reWsion 

 of the existing collections from a trained specialist wanted. I 

 believe that mammals and ants are both of the same age ; their 

 migrations took place by means of the same land connections, 

 with the difference, that winged females of ants could, easier than 

 terrestrial mammals, pass over sea-arms, being carried by winds. 



I admit that in the Oligocene epoch, after .\ustralia, .Africa 

 and South America had been cut off from a great northern 



1 C. Emerj-. " I.e Formiche dell' .■^mbr.l Stciliana net Museo Miner.i- 

 logico della R. Universita di, Bologna." {.Memor. Acctui. Boto^na[^\, vol. v- 

 I. 1891). 



- C. Emery. " Beitrage zur Kenntniss der Nord:unerik.inischen .Amcisen- 

 fauna. {Zootof^. J aiirhucltcr. Abth. f. Syst. 7 Bd. pp. 633-682, Taf. 22 ; 

 S Bd. pp. 257-360, Taf. 8. 1893-95 ) 



