Feb. 1 8, 1886] 



NATURE 



365 



We have observed also a number of short but useful 

 additions and emendations in the descriptions of various 

 of the bones, so that the present edition exceeds the last 

 by thirty-eight pages. The illustrations also have been 

 increased by the insertion of eight more woodcuts. We 

 wish to give a hearty recommendation to all students of 

 the Mammalia, to use this new edition of a book, written 

 by the anatomist who is admittedly one of the highest 

 authorities on their structure. 



Catalogue of the Fossil Mammalia in the British 

 Mu'seum. Part II. .Artiodactyla. By Richard Lydek- 

 ker, B.A. (London : Printed by order of the Trustees, 

 1885.) 



Mr. Lydekker published in January 1885 the first part of 

 the Catalogue of Fossil Mammals in the British Museum, 

 and in it he recorded the specimens belonging to the Orders 

 Primates, Chiroptera, Insectivora, Carnivora, and Ro- 

 dentia. He has rapidly followed this up by the preparation 

 of the second part, containing the sub-order Artiodactylaof 

 the great Order Ungulata. The Natural History Depart- 

 ment of the British Museum is remarkably rich in speci- 

 mens of this sub-order, and though in the Catalogue, in 

 the larger number of instances, only the briefest possible 

 description of each specimen is given, yet the volume has 

 reached 324 octavo pages. The collections, in addition 

 to those enumerated in the first part, which have furnished 

 specimens, are the Bowerbank, Layton, Sloane, and 

 Wigham collections. The author points out that he has 

 employed generic terms in a wider sense than is the case 

 with many writers. Thus he does not regard a difference 

 of one or more premolar teeth, or in the number of 

 digits, in allied forms, as a bar to generic unity, and 

 accordingly he includes the genus Eurytherium in Ano- 

 plotheriuiii. The Catalogue has been compiled with the 

 care which distinguishes the catalogues of our great 

 national Museum. 



LETTERS TO THE EDITOR 



[7 he Editor does not hold himselfresponsiblefor opinions expressed 

 by his correspondents. Neither can lie undertake to return, 

 or to correspond with the writers of, rejected manuscripts. 

 No notice is taken of anonymous communications. 



[ The Editor urgently requests correspondents to keep their letters 

 as short as possible. The pressure on his space is so great 

 that it is impossible otherwise to insure the appearance even 

 of communications containinginterestiiig and novel facts.'\ 



On " Seter," or Parallel Roads 

 I SEE to my great vexation that in my former letter on parallel 

 roads (N.'^ture, January 21, p. 26S) I have made a rather 

 hideous blunder in my English to the great disadvantage of the 

 clearness of my theory. I have used the French-German- 

 Norwegian term "rest" as signifying le restc. My theory is 

 that the last "rest," i.e. residue, of the inland ice formed a great 

 dam somewhat seawarti from the watershed ; I do not refer to 

 any "rests," i.e. reposes, in the great ablation. 



It is a fact that the glacier-shed in Central Norway was 

 situated as far as 150 kms. to the south or east of the watershed. 

 The direction of the striae and the boulder-transport renders 

 this indisputable. By the melting of the ice I now suppose 

 that the last remains must have lingered near the glacier-shed. 

 We find the last residue far to the south of the water- 

 shed. I cannot find with Mr. Melvin that this idea reverses 

 the order of Nature. The precipitation and temperature in 

 Christiania and in Trondhjem now differ very inconsiderably, 

 and the difference in height between the former glacier-shed 

 and the watershed is not very great, while the distance from 

 this to the sea is five times as great to the south as to the 

 north. 



This residue now in all valleys dams up lakes to the cols ; in 

 these lakes the terraces of gravel with laminated clay are built 

 up ; and on their shores the seter or parallel roads are formed. 

 Nothing is simpler. 



When I first got this conception I only knew the Osterdal 

 (and Lochaber) seter, but I concluded that parallel roads and 

 inland terraces were to be expected in all valleys where the 

 strias proved that the glacier-shed lay seaward to the water- 

 shed. I next found some notice of such formations in the 

 neighbouring valleys Gudbrandsdalen in Norway, and Herje- 

 dalen in Sweden. Having already finished my paper, I got 

 a dissertation of Hogbom in which, as expected, Jemtland was 

 included in my «/<? region. In my letter to Nature I further 

 inferred that parallel roads must needs exist in Swedish Lapp- 

 mark. This conclusion has also since proved to be correct. 

 Dr. Svenonius has found a sete at Sitasjaur : the correlation of 

 stri;e going upwards against the drainage with terraces and 

 parallel roads at a height corresponding with the cols. This is 

 established between 61° 40' and 68° in Scandinavia as well as at 

 Lochaber. Nowhere else are parallel roads known in Europe. 

 This local geographical distribution is perhaps the best argument 

 for my theory of lakes dammed up by the gradually diminishing 

 residue of the inland ice situated at a distance from the water- 

 shed, near the former glacier-shed. 



Mr. Melvin's theory of lateral moraines was also my original 

 working hypothesis ; but it gives no explanation of the great 

 terraces which are connected with the parallel roads, nor of 

 the laminated clay (with Desmidiefe) in the terraces as well as 

 in the sete itself Any one who has walked for kilometres on a 

 sete, smooth as a road, without any variation of the aneroid (the 

 greatest difference in Lochaber is 4 metres), will hardly be 

 able to dismiss the idea of a water-level. How Mr. Melvin 

 will account for the alternating shelves cut in the rock I cannot 

 see. Andr. M. Hansen 



University Library, Christiania 



P. S. — Err.ita in my former letter: p. 268, col. 2, line 11, for 

 " 280 kms." read " 150 kms." ; line 30 from bottom, for "till" 

 read " tell." 



Mimicry in a Neuropterous Insect 



I HAVE been much struck by a somewhat complex form of 

 mimici-y in a neuropterous insect of the genus Manlispa, which 

 would not be suspected if only a cabinet specimen were seen, 

 with the wings extended motionless, with its raptorial fore-legs 

 folded in front of the head. The insect, as I observed it, was 

 on the bare whitewashed wall of a house at Delhi, exposed to 

 the afternoon sun. ."Vs I then believed it to be a dipterous 

 insect feeding on some substance stuck to the wall, it is probable 

 that its prey, most likely the common house-fly, would be simi- 

 larly deceived, and, being attracted to the spot in hopes of 

 sharing the food, would fall a victim. 



The prothorax is curiously modified, both in colour and shape, 

 so as to resemble a proboscis, while the head and fore-legs are 

 so compactly folded that they look like some solid substance 

 adherent to the wall or stone on which the insect is resting, and 

 not part of the creature itself The mesothorax has two eye- 

 like spots shaded so as to simulate the reflections of light from 

 the compound eyes of an insect, while the markings of the 

 abdomen, seen through the transparent wings, are very dipterous 

 in character. 



The points where the Mantispa seems to fail in its likeness to 

 a fly are in the size of the prothorax, which is more massive 

 and thick than the proboscis of any fly ; there is a want of 

 prominence in the mesothorax representing the fly's head ; 

 the venation of the wings is different ; and, lastly, there are 

 apparently only four legs instead of six. 



These faulty points are seen at once on a minute inspection ; 

 but it may be imagined that it is only necessary to attract the 

 attention of a fly passing at some distance, and convey a certain 

 mental impression, which in the simple mind of a fly may not be 

 effaced till the desired object has been attained, and the victim 

 brought within reach of the Mantispa's arms. 



The resemblance between the fore-legs of the praying Mantis 

 and the same organs in Mantispa is remarkable when it is 

 remembered that the two insects belong to diflerent natural 

 orders. The fore-leg of Mantispa is the more specialised, and 

 has great lateral motion, while the edges of the femur are armed 

 with teeth slightly blunt at the tips, so that the captured insect 

 can be shifted if necessaiy. The joints in the same limb in the 

 Mantis are simple hinges, and both the femur -and tibia are 

 fringed with a double row of very sharp spines, which are neces- 

 sary to pierce and retain a hold on the thin unsubstantial wings 



