points from those of the above writer. The gall, according to him, is found in spring, the perfect insect appearing towards the end of June. This may be an earlier brood. The habit of the larva, as observed by him-social, 2-4 in a cell-is probably an effect of the larger size of the leaflets of T. pratense as compared with those of T. repens. It will be observed that on one occasion I found four inhabiting a single cell; this, however, seems very exceptional here.

III.-On Uniformity of Method in recording Natural History Observations, especially as regards Distribution and Migration; with specimen tables of a plan proposed. By Mr JOHN A. HARVIEBROWN, F.Z.S., corresponding member.

"The

The want of some definite and generally adopted method for registering in tabular form observations on Natural History, more particularly as applied to the subjects of distribution and migration, has long been felt. We have no higher authority for this statement than the author of the most important Zoo-geographical publication that has ever been published: I mean Geographical Distribution of Animals." Mr Wallace, the author, complains of the heavy task of collecting, and sifting the innumerable records of authors in various publications, in no two of which, perhaps, are they arranged upon the same plan. In summing up this part of his work, Mr Wallace says:-"Till every well-marked district, every archipelago and every important island, has all its known species of the more important groups of animals catalogued on an uniform plan, and with an uniform nomenclature, a thoroughly satisfactory account of the geographical distribution of animals will not be possible."

Whilst acting on the hint conveyed in this passage, and endeavouring to bring the matter more prominently forward, I do not do so in the belief that the plan about to be proposed, and which has lately been adopted by me in one or two papers treating of the distribution of the birds of North Russia (Annals and Mag. Nat. Hist., 1877), even approaches perfection, nor do I put it before you necessarily for imitation, but simply for the purpose of giving the subject ventilation, and of starting it for discussion. No one, I am sure, could suggest, with more certainty of being attended to, some final uniform method to be adopted, than the author just quoted, and I have not hesitated in some

measure to avail myself of the hints conveyed in the pages of his grand work, which may now be considered the common property of all workers in this branch. My object now is to agitate for an uniform method in treating of the distribution of species in minor areas, and not to consider the more general and wider aspect of the subject in relation to the great primary regions of the earth. This part of the subject stands already on a secure basis, thanks to the work of some of our leading naturalists, foremost amongst whom are the laurels due to Dr P. L. Sclater, who was the first to grasp the varied facts at his command, and mould them into the one great plan of the Zoological regions, as at present adopted by Mr Wallace, and followed by nearly all workers in Distribution;* a plan which has stood the test of discussion and criticism since its inauguration by Dr Sclater in 1856, and which appears again, with scarcely any alteration, in Mr Wallace's work.

Much may be learned by a study of Mr Wallace's methods, as shown in the tables appended to the sections treating of the different great regions; however, I humbly consider that method is even more necessary in the genesis of such work than in its final results, and that in smaller areas there are more minutiae available for record than are perhaps necessary in wider generalizations.

What we require, then, to consider may be stated thus:

"An uniform method to be established for the registration of facts connected with the distribution and migration of species in limited areas, which can be made available by field naturalists, and afterwards by cabinet naturalists, for wider generalization."

It is not my intention here to review or criticise the methods of previous authors at length, but I may be permitted shortly to speak of a few by name, referring you to personal study of their methods, for comparison and adoption, or for rejection in part or in whole.

Beginning at home, I will first instance a most excellent paper, by Mr A. G. More, "On the Distribution of Birds in Great Britain during the nesting season" (Ibis,' 1865, p. 1). As applied to a Zoologically-known area, it has much to recommend it, but it can scarcely be applied to a partially or wholly unknown * Vide Sclater (Jour. Proc. Linn. Soc., vol. ii., Zoology, p. 130); Gunther (Proc. Zool. Soc., 1858); Huxley (Proc. Zool. Soc. 1868).

SPECIES.

area, as the divisions are extremely arbitrary, and however suitable to the district under treatment, are not of sufficiently general scope to be applicable to other countries. Therefore we must look at Mr More's paper rather in the light of a practically finished work, than as a model to workers in less known areas.

The same may almost be said of another work, that by Dr. Salvadori, or rather of the tables contained in it-" Fauna d'Italia, Uccelli; Introduzione." In these the author introduces a much larger set of observations than Mr More does, and records the faunal value of each species in each of three zones, viz., Zona settentrionale, Zona centrale, and Zona meridionale, and also the insular groups, or Italia insulare; and in each of the columns allotted to these zones he indicates the species which are-stationary, s; summer residents, me; winter residents, mi; of regular passage, pr; of irregular passage, pi; of occasional or accidental occurrence, a; etc. As will be seen by the specimen given below, any of the above particulars relating to the species can easily be found by the aid. of the perpendicular columns opposite the names of the species. This plan of Salvadori's has more power of applicability than Mr More's, but the symbols or contractions admit, I think, of improvement still in that respect.

Again, M. Severtzoff, in his "Fauna of Turkestan,

s pr

* "Vertikalnoe e Gorozontalnoe Raspredalenie, Turkestanie Javotnie," Moscow, 1873. (Translation in the Ibis,' 1875, p. 96, by Mr H. E. Dresser).

Nizzardo.

Liguria

Toscana.

Romano.

Napoletano.

Sicilia.

Sardegna.

Malta.

taken as the type of another class of writers, who introduce yet another useful element into their tables of distribution. He gives tables of both horizontal and vertical distribution, which are matters of the utmost importance, and which should never be lost sight of.

These, and many other examples of modes of registration by different authors, which I might give you, do not, however, appear to me to supply the desideratum of general applicability throughout. At the same time, I am inclined to consider Mr Severtzoff's divisions to be the most easily applicable to all countries, whatever their physical aspects or contours may be, but with certain modifications. Thus, while the four horizontal divisions of Turkestan adopted by Mr Severtzoff may be most suitable to that country-viz., north-west, north-east, south-west, and south-east,it might be desirable to introduce others in countries of greater magnitude or different contour, or more varied physical aspects, as Salvadori has done with his "Italia insulare," or, as will be seen later, I have done in my treatment of the Distribution of Birds in North Russia. Generally speaking, however, I am inclined to adopt this method of division, and therefore I would suggest :

1st, That the horizontal divisions of a limited area or country, Zoologically-known or Zoologically-unknown, be determined by their relative positions according to the compass, with such additional districts as may be considered convenient, such as central or insular.

2d, That vertical divisions be determined in each country by their altitudes above the sea; and that they be further individualized by having appropriate names bestowed upon them, according to the physical aspects peculiar to each.

Examples.-In Norway I would, for instance, propose four vertical districts (or five, if we include the adjoining fjords and seas), viz. :-(1) Littoral, or sea and shore; (2) Pine tract, or sealevel up to upper limit of pine growth; (3) Birch tract, or to upper limit of birch growth; (4) Fjeld tract, or Mountain tract, to summits of about 8000 feet; and (5) Insular, or outlying islands. But this last, in the case of Norway, is perhaps un

necessary.

Transylvania, again, I would divide as follows:-(1) Field lands, or plains and valleys; (2) Foot hills, or oak tracts, say to an

elevation of 2000 feet, or about 800 feet above the high level of the Field lands; (3) Beech tracts, or lower mountain slopes, say to an elevation of about 3000 feet; (4) Pine tracts, or central mountain slopes, 4000 to 5000 feet; and (5) Lichen tracts, or upper mountain slopes and summits, 5000 feet to about 8000 feet.

So much, then, for the division of an area into districts.

As regards migratory movements, I think, as far as possible, all records connected with this subject should be kept separate from those of distribution. I am aware that it is not possible altogether to ignore migration in a treatment of distribution, but we can, to a certain extent, give such records a secondary importance, by recording simply the occurrence of such species and no more, leaving for a separate table of migratory movements the entry of more detail.

We now come to the use of symbols as a means of expressing the faunal value of the species in the columns.

Professor Newton of Cambridge, in a Paper in the "Transactions of the Norfolk and Norwich Natural History Society," (1870-1, p. 24), has explained a method of registering Natural History observations, which his brother and himself followed out during a period of ten years' residence at Elvéden, in Norfolk. By those who are desirous of keeping exact daily records of migratory movements, I believe no better plan can be followed, and I would recommend to such a careful study of his article. Professor Newton tells us:- "However good an out-door naturalist any man may be, let him adopt the plan here described, and he will become a better one" (p. 29), as the keeping of such records "ensured continuously our careful attention, forcing us to take notice of everything that caught the eye." I will only further say here that Professor Newton recommends what I may call a progressive system of symbols, having its chief recommendation in its "variability, combined with simplicity," and "a capability of expressing various meanings, especially in combination, and its freedom from complication." Bearing this advice in view, I would then suggest for approval the following code of symbols for recording observations on Distribution, and which can be used either in the field for daily work, or in the study for bringing into workable order the records of previous authors.