at the latter from the S.E., so that it was thought the columns were not identical. At Angers a similar thing was observed on June 10th, the direction being from E, to W, against the wind,

which was not of much force. It was estimated that the number passing along a single street in one hour was from 40 to 50,000. At Würtemberg, from June 1st to 8th, an incessant migration from S. and W. towards N.E. and E. was remarked. In the

Canton Zurich, on June 7th, an immense swarm moved from S.W. to N.E., principally from 3 to 5 p.m.; this flight went persistently in one direction, only changing temporarily to avoid houses and trees. At Steyer in Upper Austria, on June 11th, surprising numbers flew ceaselessly from S.W. to N.E.; between 1 and 2 p.m. 90 to 110 per minute were counted in a space of about 100 paces, and the swarm was estimated to have been above rather than below 1,000,000."

These particulars, amongst others given by Mr. McLachlan, will suffice to show the very extraordinary nature of the migration of Vanessa cardui observed in that particular year.

Commenting upon the facts, Mr. McLachlan says (l. c.):"There can, I think, be little doubt that all the swarms consisted of individuals that had hybernated; there can also be little doubt they were migrating, and that the column had become dispersed before remnants of them reached our shores, and other parts of the North of Europe. . . . Whence came they? Were they all bred the previous autumn? or is it possible the insect may be able to rest quiescent in the perfect state over a series of years until the accumulated numbers simultaneously wake up? The whole subject is surrounded with difficulties."

When exhibiting the Ceylonese specimens, forwarded by Mr. Mann, at a meeting of the Linnean Society on the 6th of June last, I remarked that these curious flights of butterflies could not be regarded as quite analogous to the migration of birds, which travel in opposite directions in spring and autumn; for the insects travel only in one direction and do not return, vast numbers perishing en route. The phenomenon resembles rather what has been observed of Lemmings,* Locusts,t and Dragonflies, and may be explained as a sudden exodus from the birth

Crotch, Linn. Soc. Journ. xiii. pp. 27-34 (1878); additional note, p. 83. Further remarks, pp. 157-160 pl. xiii. Collett, tom. cit., pp. 327-334. Somerville, Proc. Zool. Soc., 1891, pp. 655-658.

See Layard, 'Birds of South Africa,' pp. 291, 314; and Andersson, 'Birds of Damaraland,' p. 264.

Several of the larger species of Libellula occasionally migrate, but the phenomenon is of rare occurrence, and the circumstance which brings about such an uncommonly numerous development of the perfect insect must be very peculiar. See Weissenborn, on a remarkable migration of Libellula depressa, Mag. Nat. Hist., n. s., vol. iii. p. 516; and Hudson, on "Dragonfly Storms" (composed chiefly of Eschna bonariensis), The Naturalist in La Plata,' pp. 130-134. See also Gätke, on the migrations of Libellula quadripunctata, Heligoland, p. 89,

place, leading to a compensating reduction of the species after a season exceptionally favourable to its increase. In this view Col. Swinhoe, whose extensive acquaintance with the Lepidoptera of India and Ceylon is well known, has expressed his concurrence. But since these remarks were made, some additional and most interesting information has come to light, and from an unexpected source. Herr Gätke, who for fifty years has devoted himself to a study of the migratory birds which visit Heligoland, has published, in a bulky volume, the results of his observations, and an English translation of the German text has just been issued. In this volume will be found some extremely interesting observations on the migration of butterflies, from which the following is an extract:—

"That Lepidoptera during their more extended flights are subject to the same meteorological influences as birds, I have been convinced by the experience of many nights in July, during which I have caught numerous nocturnal Lepidoptera not belonging to our insect-fauna, the weather of these nights being invariably such that, if it had occurred a few weeks later, it would have conducted hither innumerable Wheatears. It has, in fact, occurred repeatedly that Lepidoptera, especially night-flying species, have passed over this island in countless swarms at the time of powerful bird-migrations. Thus, during the night of the 25th of October, 1872, thousands of Hybernia defoliaria, mixed with smaller numbers of H. aurantiaria, passed over the island in the company of large numbers of Larks. And again, in the night of the 11th of October, 1883, during which an unusually strong migration of all the species of birds due at that time took place, this was accompanied by the appearance of very large swarms of the same species of Hybernia.

"White Cabbage Butterflies, the Black Arches, Psilura monacha, and also Libellulidae (Libellula quadripunctata) have been seen to pass here in migratory flights of astonishing proportions, though even these do not come up to those of Plusia gamma, which on repeated occasions have occurred in numbers of which it would be quite impossible to form any

Heligoland as an Ornithological Observatory: the result of fifty years' experience by Heinrich Gätke.' Translated by Rudolph Rosenstock. Roy. 8vo. Edinburgh: David Douglas. 1895.

conception. Thus, during the nights from the 15th to the 19th of August, under the favourable conditions of a south-easterly wind and fair weather, a considerable migration of birds took place. On each of these nights, from 11 p.m. to 3 a.m., the small Noctuæ above referred to were seen at the lighthouse, passing from east to west in undiminishing numbers, like the flakes of a dense snowstorm. These small creatures also manage to cross the North Sea in safety, for they often arrive on the east coast of England suddenly, and in such remarkable numbers that we can only believe them to be immigrants. In fact, as my friend Mr. John Cordeaux informs me, an enormous accumulation of these insects actually took place at a time corresponding to that of the above observations." (P. 117.)

Further on Herr Gätke writes:

"Both the theory of inheritance as well as that of tradition are quite untenable in regard to the periodical migratory phenomena of other animals endowed with the power of flight, such as beetles and nocturnal Lepidoptera. Among the latter the migrations of Plusia gamma furnish in Heligoland excellent material for observation. During their autumn migrations these small moths travel from Slesvick-Holstein to England, across the North Sea, a breadth of water of four hundred miles in extent. They pass this island in enormous swarms, resembling, as seen from the lighthouse, a dense snowstorm driven by a light breeze. Thus, according to an entry in my journal for 1882, on the night from the 15th to the 16th of August, with a very light south wind, a powerful migration of birds occurred:- From 11 p.m. to 3 a.m. millions of Plusia gamma were travelling from east to west, like a dense snowstorm.' 'Again, on the nights of the 16th, 17th, and 18th, large numbers of P. gamma passed the island, the migration commencing each evening at 11 o'clock. On the 19th the wind was south-east, the weather fine and calm. In the evening the sky became over-cast, and a strong migration of birds took place. From 11 p.m. until 2 a.m. thousands of P. gamma were again seen.' A thunderstorm, with high winds, subsequently put an end to the migration. These little insects also follow an east-to-west course of migration, and they adhere to it with as much steadiness and precision as the different migratory hosts of birds which are observed here. That they, too, accomplish their journey in safety is shown by the enormous swarms of them which frequently cover

the east coast of England, and which can only be explained as the result of an immigration. Besides P. gamma large numbers of Gastropacha neustria, Agrotis graminis, and other species, are represented in such migratory swarms. It has been suggested that these insects are attracted by the light of the lighthouse, and that, consequently, it is only around the latter that they are seen in such quantities; this, however, is contradicted by the migrations of Hybernia defoliaria and H. aurantiaria, these insects sometimes making their appearance during strong migrations of Larks in October, when large numbers of them may be found in the course of the night, as well as on the following morning, from one end of the island to the other. Now it is quite impossible that these moths should be able to collect experiences of any kind during this single migration of their life, which, moreover, is performed in the darkness of night across a wide expanse of water; and even if they did these would be perfectly useless, for these migrants die shortly after their autumn migration without having produced further offspring to which they could commit their experiences, either by hereditary transmission or by personal instruction."

On the subject of migrating Dragonflies, Herr Gütke remarks:

"Another very peculiar phenomenon, also intimately connected with thunder-storms, is the regular but temporary appearance, in millions, of the large Dragon-fly (Libellula quadripunctata) before such disturbances. Countless swarms of these insects make their appearance all of a sudden during the calm sultry hours preceding the catastrophe, while thunder-clouds gather on the horizon, and, heaped upon each other, project into the blue ether beyond, like so many giant mountains of snow.

"The direction whence these insects proceed cannot be ascertained, nor do they arrive in swarms or companies, but solitary individuals or scattered groups probably congregate on the spot in one vast throng. The assembling individuals or groups must, however, follow each other in very rapid succession, for in a short time the face of the cliff, still illumined by the sun, all the buildings, hedges, and dry twigs on the island, are covered with them. Nor is it necessary for the occurrence of this phenomenon that the storm should actually discharge itself over Heligoland, or even in its immediate neighbourhood, but