H Y M H Y M 571 Mason s hygrometer, but more generally known as the dry and wet bulb hygrometer, which has the strong recom mendation of being self-acting, and requiring only the read ings of the two thermometers in making air observations. This hygrometer (fig. 3) consists of two thermometers a and b similar to each other in all re spects except that one of them has a piece of muslin tied closely over the surface of its bulb c, and kept constantly wet by a few threads of cotton which con nect it with the water in the vessel d. The water then which rises from the vessel by capil lary attraction spreads over the muslin, and evaporates from its surface with more or less rapidity according to the dryness or moist- ness of the air ; and the greater the dryness of the air the greater is the difference between the observed readings of the dry and the wet thermometers. The formula for deducing the hygrometric state of the air from these two observations has been investigated by Professor Apjohn (Trans. Roy. Irish A cad., vol. xvii.), and has been already de scribed (see ATMOSPHERE, vol. iii. p. 32). As it is very fn n 90 1 1: 90 80 So 70 70 &c CO 5o 50 46 4o Jo ? 2f- 20 10 ^_ 10 13 p tip ( ) ^ =^ ,7 i Fro. 3. Dry and Wet Bulb Hygrometer. troublesome to go through the calculations for each fresh observation, tables have been prepared which give the dew- point by inspection. The best of these tables in English measures are those of Glaisher, fifth edition, constructed empirically from direct experiments carried on at Green wich, combined with Regnault s last revised tables relat ing to aqueous vapour. There can be but one opinion as to the great service rendered to meteorology by Glaisher in the preparation of these tables, which give results approximately correct for high and moderate humidities and for situations at no great height above the sea ; in other words, they may be regarded as accurate for at least such conditions as are presented by the climates of the British Islands. They are, however, insufficient, owing to the com paratively large errors attending their use, for the reduction of observations made in elevated situations and in such arid states of the atmosphere as are of frequent occurrence in India and South Africa. The preparation of such tables remains still a serious desideratum in meteorology; and another desideratum equally important is the introduction of a simple, handy, and accurate method of observing the hygrometry of the air when its temperature descends below the freezing-point of water, some method which would in volve only a minimum of manipulative skill and trouble in making an observation. See ATMOSPHERE. (A. B.) HYMEN, or HYMEN^US, was originally the name of the song sung at marriages among the Greeks. As usual the name gradually produced the idea of an actual person whose adventures gave rise to the custom of this song. He occurs often in association with Linus and lalemus, who represent similar personifications, and is generally called a son of Apollo and a Muse. In Attic legend he was a beautiful youth who, being in love with a girl, followed her in a procession to Eleusis disguised as a woman, and saved the whole band from pirates. As reward he obtained the girl in marriage, and his happy married life caused him ever afterwards to be invoked in marriage songs. At other times the tale is of an opposite character ; Hymen was unfortunate either by dying on his marriage day or in some other way, and ho was invoked to propitiate him and avert a similar fate. He occurs often in the train of Aphrodite, along with Eros, Himerus, Fothus, &c. According to Orphic legend he was restored to life by Asclepius. HYMENOPTERA (the Piezata of Fabricius, HaiUfluger or Aderfliiger in German), an order of Insecta (so named from their wings being joined, as hereafter described) containing the insects commonly called bees, wasps,, ants, ichneumons, gall-flies, saw-flies, and others less known which have received no English names. The main charac- General teristics of the order are these : the possession of four cn *racte wings, of which the anterior are always larger than the isties> posterior, always of the same texture, and mostly with nervures arranged in regular patterns ; a dense hard skin, smooth, shining, or very hairy ; a mouth always provided with mandibles adapted for biting, though the other mouth parts may be so modified as to serve for mastication, or for the sucking, or rather lapping, of liquids. The female is provided with an anal instrument connected with oviposi- tion, and sometimes serving for defence, in which case it is in communication with a poison gland. They undergo a regular transformation, and have larvae provided with legs on the thorax and abdomen, or on the former part only, or (as is more often the case) entirely footless. The wings have few nervures, and may be even entirely Wings, devoid of them; when present they proceed from the base of the wing, or from the costa (the part which bounds the wing in front) towards the apex, which they may or may not reach. Connected with the lateral nervures are others (recurrent nervures) which unite them together, and form in this way regular cellules. The first two lateral nervures (those nearest the top) are the most important, and are called the "marginal" and " submarginal" respectively. The cellules which they form by means of the cross or re current nervures are called the "marginal" and "sub- marginal cellules." Not unfrequently there is, towards the apical third of the costa, a thickened spot (considered by some to act as a counterpoise when the insect is flying) termed the "stigma" or " pterostigma." It is not always present, however, and when present may be very small, or, as in Pachylostica, very large and projecting. In a similar way the hind wing may contain lateral and cross nervures, but they are fewer in number and in importance, and may be entirely absent, although present in the larger pair. The arrangement of the nervures in the Hymenoptera undergoes, in the various families, great diversity ; and, what is of great importance in classification, their form, in the different families, and even genera, is, within certain limits, remarkably constant. On this account great attention has been paid to them ; and each nervure and cellule has received a distinct name. The use of the nervures and cellules does not lead to the formation of artificial groups ; for we find that the existence of a particular arrangement of the nervures in a hymenopterous insect denotes the pre sence of other characters. Each family indeed has its own form of wing, as will be seen from the accompanying figures. The relative value, however, of the ncuration in classifica tion is not always the same. It is of much greater import ance, for instance, with saw-flies than with bees. The wings are usually shorter than the body, and may be so short as to be useless for flight ; they may be even entirely absent. Apterous species are found in almost all the fami lies. In most cases it is the females only which are thus deprived of the power of flight. But the opposite of this may exist ; as, for instance, with a curious species of Chalcididce which lives in the nest of bees as a parasite. "With ants, again, the neuters are always wingless ; and the females lose their wings when they commence the formation
of a colony. It is worthy of remark, too, that some Chalci-
