remorse at sea. This is a phase of the seaman's life which is worthy of the attention of the philanthropist indeed. It is Jack ashore, and not Jack at sea, that any man, whoever he is, who would improve the sailor, must look after. We have more faith in reformation that might be effected by the influence of lads leaving home with the recollection of mothers and sisters fresh in their hearts, and with a love for home and home ties superior to the attractions offered by the feverish debauch so persistently placed before them under the licensing laws, and by the crimp and his coadjutor the harlot, than we have in any movement, however sensational, founded in an attack on British ships and British shipowners. A means of spending time in port, which shall be a happy medium between a barrack home and revival prayer meeting on the one hand, and a drinking shop and brothel on the other hand, would do more to improve the moral condition of our seamen than all the legislation that will ever puzzle members of Parliament, or “meddle and muddle" with ships. The education of boys for the sea service has not yet obtained the attention it deserves, and save the exceptional cases of one or two eminent shipowners, and of one or two earnest men attached to seamen's missions, the real wants, the shore wants of seamen are neglected for the more fanciful and more notorious agitations respecting seamen afloat.

It is unnecessary to carry our remarks further. The return seems to us to be a record of vice, dissipation, idleness, ignorance, and roguery, even if presented in the most favourable light for the men; but even then we are happy and proud to say, that so small is the proportion of men included in this return when compared with the great body of seamen, that it leaves a splendid and immense majority whom every Englishman has a right to claim. with pride as fellow subjects.

AN INQUIRY REGARDING THE CAUSES OF THE GENERAL CIRCULATION IN THE ATMOSTHERE.

FEEL much honoured by the notice which Herr Buys Ballot, the distinguished Dutch Meteorologist, has taken of my former article on "Gradients and Currents," and I desire to express my indebtedness to him for the valuable suggestions which he offers for consideration.

In the remarks on circular waves my aim was not to present an exhaustive view of their interactions, but only to indicate and illustrate in a very general manner my conviction that the fundamental element in the varieties of pressure experienced in the atmosphere is the simple wave form with its advancing and receding side, and that it is by the combination of such rotating waves that the resulting depressions or elevations are formed.

Since writing the article the following illustration suggested itself, exhibiting in a simple way the development of a cyclonic movement between two rotating waves.

If A and B, representing two hemispheres impinging upon a central one C, having its flat side uppermost, be made to rotate

with watch hands, the central C partaking of their movement will rotate in the reverse direction.

A depression being a negative quantity and a wave a positive

à priori, one would imagine the latter a more instructive study than the former. However, I quite agree with the opinion that neither should be neglected, and that the greatest advance may be expected to be made by considering both, and their relations to each other. I hope (with the Editor's kind permission), when I come to consider the formation of these local gradients, to offer a few practical remarks in the lines of inquiry marked out by Herr Buys Ballot. Until we can form some notion of the influences to which these waves are subjected, we can hardly reason with accuracy regarding their movement, even with the aid of the most carefully constructed charts, for under a present influence they may be proceeding in a certain direction, and again in a few hours receiving a new impulse, their form and direction may be completely modified. An attempt is made in the Table of Currents to point out the times and nature of these influences, and in treating of local differences of pressure it will be my endeavour to state the principles upon which it is constructed.

Astronomers are well aware of the value of knowing the best times and places for making their observations, and it is hardly necessary to urge on meteorologists the importance of being acquainted, even approximately, with the time and character of an impending change; as by repeated observation its real nature may be ascertained.

With regard to the extension of the Table it should, I think, pretty nearly suffice for any place in the north temperate zone, reckoning the time at such place by the sun, and making a slight allowance for variation of latitude. In the extreme north, east should be substituted for west in the columns headed "General Direction" and "Force."

But as in my opinion the key to the local circulation as distinguished from the general is to be found in the latter, it is necessary in the first place to examine the causes which effect it, eliminating as far as possible the local irregularities with which it is overlaid.

In the former article we saw that to produce the trade winds. there must be two great gradients within the tropics, having their lowest parts near the equator, and their highest in the belts of calms

and variables to the north and south respectively. This theoretical deduction is verified by the results of observation, which show that the mean of the barometer is higher at the tropics than at the equator; in other words, there is an almost constant deficiency of pressure in the equatorial regions.

Fig. 2 (b) gives a representation of a vertical section of these two gradients. A and B representing regions of higher barometer

north and south of E, the equator a region of lower barometer, A E being the gradient for north-easterly winds, and B E for south-easterly.

If, for the sake of illustration, we suppose (Fig. 2, a) the barometer to stand at a uniform height within the region in question, then to produce these gradients the barometer must fall at E'. A depression being formed there, a motion of the air from A and B will take place towards it, causing the barometer to rise at A and B, both actions producing the same result. It will be seen further on, that this depression is followed by a compression, as it may be called, which will have the effect of causing the barometer to rise at E, and that again will cause a fall at A and B. This will render the gradients less steep. But this compression being succeeded by another depression, the same variation will again

occur.

Whatever difference of opinion there may be as to the causes at work, it may be safely asserted that there is none as to the region in which they are most effectively in operation, and that is in the equatorial regions. We are thus driven to the conclusion that these gradients are mainly due to a fall of the barometer or the formation of a depression in the vicinity of the equator.

That the general circulation is owing to a depression, followed

by a compression moving from east to west, may be shown

Let P (Fig. 3) represent a particle or body of air moving towards the centre of a depression D-the barometer being higher at P than at D. When the particle has reached the position P1 the depression moving towards the west is now at D', and the particle will therefore move towards it into the position P2, but the depression being now at D2, the particle will move to P3; in plain terms, the motion of the depression will cause the body of air to describe a semicircle-moving from west through north to east.

The compression by which, according to the law of inertia, this depression must be followed, will complete the circulation of the body of air P. For if we consider that while the particle is in the position P3 the following compression is in the position D, the barometer being now higher at D than at P3 the particle will move into the position P. As the compression proceeds westwards to D1 and D the particle will move towards P5 and at last regain P, its original position. In short it will describe a semicircle moving from east through south to west.

In a similar manner it might be easily shown that a depression moving westwards, followed by a compression, will cause the air south of the equator to rotate in the reverse direction.

This result is in strict agreement with what was observed in the experiment with the water in the basin-the circulation being clearly the effect of a hollow, followed by an elevation.

It may now be naturally asked, is this experimental and theoretical