of the cirrus down to those nearest the earth where the fracto-cumulus is produced, as the vapor of water passes from the state of frozen particles to that of aqueous globules, or vice versa. However, the palliocumulus, which serves as a transition from the two types and their derivatives, is found a little more elevated than the cumulus.

I have thought it proper to modify Forster's nomenclature by substi tuting names more in harmony with the form and nature of clouds. I give, in continuation, the old and the new classification:

With the exception of cirrus, whose name curl-cloud approaches nearest the form of that cloud, all the determinations have been changed. The pallio-cumulus replaces the nimbus, also named rain-cloud.

I.-CIRRUS, (HOWARD.)

Curl-cloud-cirrus, so named by Howard, (the "cat's-tail" of sailors, illustration No. I, figs. 1, 2,) is composed of filaments which resemble a twisted tuft of curled hair, (illustration No. I, figs. 3, 4,) plumage, (fig. 5,) the flowing tail of a horse, (illustration No. II, fig. 6,) or a fine pencil, (illustration No. III, fig. 7;) at other times are disposed in long, straight bands, parallel to each other, or divergent, palmated, or like a herringbone or vertebral column, (illustration No. IV, fig. 8,) their greater axis oriented according to the sailing of the cloud and the direction of the wind at that altitude, which soon makes itself felt at the surface of the earth. When it forms two or more systems of straight, parallel bands, by an effect of perspective they appear to diverge from their point of departure at the horizon, and to converge toward the point of the horizon opposite, as do the rays of the rising and setting sun.

The cirrus is always white-sometimes brilliant, sometimes pearlydull. The earliest and latest reflections of the solar rays upon these clouds color them with a delicate rosy tint, more or less intense, accord, ing to their density. Their movement is exceedingly slow, and their altitude is not less than 10,000 yards, (more than six and a quarter miles.) These clouds are the highest, apparently, slowest, most rarified, most variable in their forms, and the most extended. The appearance or disappearance of cirrus proclaims the end or the commencement of good weather. The barometer sinks and then rises, all the accompanying meteorological phenomena undergoing a similar change. We quote from Howard:

They are first indicated by a few threads penciled, as it were, on the sky. These increase in length, and new ones are in the mean time added to them. Often the firstformed threads serve as stems to support numerous branches, which in their turn give

rise to others. The increase is sometimes perfectly indeterminate; at others, it has a very decided direction. Thus the first few threads being once formed, the remainder will be propagated in one or more directions laterally, or obliquely upward or downward, the direction being often the same in a great number of clouds visible at the same time; for the oblique, descending tufts appear to converge toward a point in the horizon, and the long, straight streaks to meet in opposite points therein; which is the optical effect of parallel extension. The upward direction of the fibers or tufts of this cloud is found to be a decided indication of rain; the downward as decidedly indicates fair weather. Their duration is uncertain, varying from a few minutes after the first appearance to an extent of many hours, and even days. It is long when they appear alone and at great heights, and shorter when they are formed lower and in the vicinity of other clouds. This modification, although in appearance almost motionless, is intimately connected with the variable motions of the atmosphere. Considering that clouds of this kind have long been deemed a prognostic of wind, it is extraordinary that the nature of this connection should not have been more studied, as the knowledge of it might have been productive of useful results. In fair weather, with light, variable breezes, the sky is seldom quite clear of small groups of the oblique cirrus, which frequently come on from the leeward, and the direction of their increase is to windward. Continued wet weather is attended with horizontal sheets of this cloud, which subside quickly and pass into the cirro-stratus. Before storms they appear lower and denser, and usually in the quarter opposite to that from which the storm arises. Steady, high winds are also preceded and attended by streaks running across the sky in the direction they blow in.

II. CIRRO-STRATUS, (HOWARD.)

Thread-cloud, (illustrations Nos. V and VI.)-Howard's cirro-stratus is distinguished from the pure cirrus by its filaments being smaller, more compact, more ramified, and, so to say, completely stratified. They are lower and more dense, for the sun's rays often pierce them with difficulty. Their whitish tint is clearer, and it becomes rose-color in similar circumstances. Their motion is a little more rapid. When at the horizon, we only seeing the vertical projection, they take the appearance of a long and very narrow band. According to Howard, "this cloud appears to result from the subsidence of the fibers of the cirrus to a horizontal position, at the same time that they approach each other laterally. The form and relative position, when seen in the distance, frequently give the idea of shoals of fish. Yet in this, as in other instances, the structure must be attended to rather than the form, which varies much, presenting at times the appearance of parallel bars, or interwoven streaks like the grain of polished wood. It is thick in the middle, and attenuated toward the edge. The distinct appearance of a cirrus, however, does not always precede the production of this and the last modification. The cirro-stratus precedes wind and rain, the near or distant approach of which may sometimes be estimated from its greater or less abundance and permanence. It is almost always to be seen in the intervals of storms. Sometimes this and the cirro-cumulus appear together in the sky, and even alternate with each other in the same cloud, when the different evolutions which ensue are a curious spectacle; and a judg ment may be formed of the weather likely to ensue by observing which modification finally prevails. The cirro-stratus is the modification which