the rock to atmospheric action by collecting the surface-crust also. The difference of colour between such crusts and the interior is often striking, as may be seen in brown clay-blocks with blue cores, or in the blue-grey "felstones" of Wales, which weather to a porcellanous white.

Fossils may often be gathered in a good washed-out condition on the loosely coherent banks immediately below the outcrop of the beds in which they properly occur; but no pains must be spared in collecting from the horizons themselves, although the specimens thus obtained may emerge fractured and otherwise obscured. The assemblage of forms should be fairly represented in a collection, since it is well known that the mere fact of the occurrence of a particular species does not necessarily mark a

zone.

Curious errors of locality often arise, which are due to the indefatigability of man. Thus remarkable rocks are carried for long distances to decorate the window-sills of cottages, and afterwards become cast out upon the hills to puzzle the wandering observer. Similarly the glassy slags of long-forgotten furnaces have again and again been produced as evidence that volcanic rocks occur in such and such an area. As to the shells of recent edible molluscs, especially the oyster, no locality seems too anomalous for their discovery.

The labelling of specimens in the field may be done by attaching numbers to them on strongly gummed labels, and describing them, with corresponding numbers, in the note-book. Each specimen should be wrapped in newspaper to prevent friction with its fellows in the bag. Hence a less cumbrous, though at the same time less neat method, is to write all the description on some part of the plain white edge of the wrapping-paper; this should be folded in so as to escape tearing, and should be always kept as carefully as the specimen itself. It will form, in fact, when subsequently torn off on unpacking each specimen, the "original label" drawn up on the spot to which all future reference must be made. Experience shows that specimens thus labelled on infolds of their wrappers may be sent thousands of miles by rail or sea without any risk of confusion or loss of the observations recorded. Dr. Blanford advises travellers in tropical countries to poison their labels by washing in a weak solution of corrosive sublimate, to prevent their being destroyed by mites and insects. (See the admirable series of papers entitled Hints to Travellers, published by the Royal Geographical Society, 7th ed., p. 382.)

Charles Darwin suggested from his experience that during a Voyage the specimens from distinct localities should have their wrapping-papers marked externally with characteristic signs, a practice of great assistance during unpacking.*

The original label, or the note corresponding to it, should be as full as possible, and may even contain cross-references to specimens collected in the same series. If the locality is an unknown one, as in long traverses and in wilder regions, the hour and day of finding should be noted in each case. Even if this goes on from day to day during the passage of an expedition, some idea can be gained respecting the relative positions of the places studied along the route.

The practice of noting the day, month, and even year, on the label of every specimen, is, indeed, of continual use in after reference; and in all careful study of an area the hour of the day is of assistance. These minute facts, like so many others, are not, however, for publication. In research-work the duty of the observer is to separate the important facts from the mass of material that has been gathered on the chance of its proving of importance. On the spot everything must be noted; later studies weed out the fundamental from the trivial. The young writer who refers to his specimens by number, and describes them each in detail, has either lost sight of their field-relations, or is working on "drifted" material in the darkness of cabinets and

museums.

The transport of geological specimens to their destination is seldom a matter of difficulty, owing to their non-perishable character. In most places cheap sacks or bags are obtainable, and these prevent the specimens from shaking on one another, as they may possibly do in a partially filled box. The bag should be of stout fibre, and should be sewn over with strong string just above the specimens. Several such bags may be sent on beforehand to the area of work, since there is often a difficulty in a small town in procuring a box of sufficient strength and of convenient size.

Fossils travel better in a wooden box, unless each can be enclosed in a small box of its own. Small and delicate specimens may be separated in stout corked glass tubes, wrapped round with paper several times; others may be temporarily glued to the bottom of chip-boxes before packing. Where wool is used, it is important to first wrap the specimen in tissue-paper, since the fibres of the wool, if in direct contact, remain upon the specimen for years.

Admiralty Manual of Scientific Enquiry (1859), p. 272.

The development of the parcel post has greatly facilitated thetransport of small series of specimens from foreign countries. It should be noted that in Italy a linen cover is required to parcels, paper alone being inadmissible; hence here, and elsewhere as a safeguard, a few calico bags to enclose the series sent. form a handy addition to one's equipment.

PART II.

THE EXAMINATION OF MINERALS.

"La Minéralogie, étant une branche de l'histoire naturelle, les mêmes principes qui dirigent les naturalistes en général doivent diriger aussi le minéralogiste. Il cherche a connoître et à apprécier la place que les divers espèces de minéraux tiennent dans cet ensemble d'êtres qu'on nomme la nature, et à s'instruire du rôle qu'ils y jouent."—ALEXANDRE BRONGNIART, Traité de Minéralogie, 1807.

CHAPTER III.

ON THE OCCURRENCE AND SOME PHYSICAL CHARACTERS OF MINERALS.

A. Mode of Occurrence.-The relation of the mineral specimen to its surroundings should in all cases be observed prior to its extraction. Its occurrence in veins or diffused through a rockmass, in concretionary forms or in well-developed crystals, its deposition upon earlier-formed constituents, or its inclusion in other substances that have aggregated round it- these are a few of the many points that may help in its final determination. If it appears to be a product of alteration, search should be made for examples of the mineral or minerals from which it may reasonably have been derived. In the case of a substance of especial interest or of commercial importance, a rough sketch or plan of the spot made in the field will often refresh the memory and assist description when the details come to be worked out later.

B. Extraction. The modes of extracting particular minerals from the mixed or massive aggregates known as rocks will be more conveniently treated of in connexion with the rocks themselves. The mineral particle, whether crystallised or not, having been isolated from its matrix, some one or all of the following methods of examination may be applied with a view

to its determination. The test of hardness, and some observations on form and cleavage, may often be employed without its removal from its surroundings; similarly, the optical tests described are far more commonly applied to minerals occurring haphazard in rock-sections than to preparations cut in known directions from specially extracted specimens.

C. Colour and Lustre Transparency or Opacity.--It is unnecessary to remind any worker among minerals of the variation of colour in one and the same species. The ores of the heavy metals are by far the most constant in their colouring; but even here the phosphates, carbonates, &c., may assume very deceptive tints. Similarly, a mineral may at times be transparent, at others apparently opaque. Sometimes, however, as in the blacker varieties of zinc blende, a small chip or two flaked off will reveal the true translucent character of the more typical mineral. The blue and blue-green colours of vivianite will similarly become visible when flakes broken from dark crystals are held up in forceps to the light. Some minerals, on the other hand, such as magnetite, are opaque even in the thinnest fragments; and this property becomes accordingly useful in their identification.

The lustre of the faces of crystals or cleavage-planes is often of service, though these are liable to be dulled by filmy products of alteration. The lustres recognised by experts are given in all works on mineralogy.

D. Streak. The streak of a mineral, i.e., the colour of its powder, can very often be observed by scraping a rough edge of the specimen with an old but clean knife, and spreading out the little fragments, under pressure of the blade, upon white paper. A refinement is to use a slab of unglazed porcelain or the side of a mortar, across which the specimen is drawn; the coloured lines thus given by different minerals may be produced closely side by side, and comparison becomes very easy. The specimens must be free from surface-films and decomposition-products. While most rock-forming minerals yield white or colourless streaks, the results given by many sulphides and oxides of the heavy metals are eminently useful and characteristic, especially when known specimens are ready at hand for comparison. It is scarcely necessary to mention the red streak of specular iron or hæmatite, the orange-brown of limonite, the grey of galena, the purple-red of pyrargyrite, or the browner red of cuprite, as familiar and practical examples.

E. External Form. - The pocket-lens will aid considerably in examining the crystalline form of minerals that have consolidated