16 EEV, A. E. EATON ON EECENT EPIIEMEEID^ OE MATELIES. 



cephalic, issuing from tlie basal joinings of the upper maxillfe. Jolia, in addition, has a 

 branchial tuft at the insertion of each of the fore legs. The abdominal pairs are usually- 

 all exposed, and are carried diversely in different genera ; their form, proportions, and 

 substance likewise vary more or less, not only with their serial position in the same 

 insect, but also with the genus. In Bcetisca and Frosopistoma all of them are concealed 

 by the shield already described in connexion with the thorax ; in other cases one pair is 

 enlarged and thickened so as to resemble elytra covering the pairs posterior to it 

 (Pis. XL.-XLII.). Several fossorial nymphs dispose their tracheal branchiae in an 

 arch over their backs ; in other kinds they are decumbent upon the dorsum ; in others 

 they are directed outwards and backwards from, or are held at right angles with, the 

 sides of the body ; in Uliithrogena some are deflected, and two pairs underlie the venter. 

 As to their forms, when foliaceous they may be subquadrangular, subrotund, oval, 

 ovate, sj)atnlatc, lanceolate, or linear, with tlieir margin entire or fringed, or in part 

 eroso-creuate ; they may be digitate, pinnatisect, laciuiate, or dissecto-fimlnnate ; and in 

 many of these alternatives they may be single or binate, plane or conduplicate. Some- 

 times they are compounded of dissimilar elements, as when a tracheal branchia consists 

 of a foliaceous membrane, furnished at its base with an exjilanate or fasciculated tuft of 

 simple or branched filaments (Pis. LIV.-LXII.); or as in Ephcmerella and its kindred, 

 where each is composed of a coriaceous lamina sheltering a binate appendage of imbri- 

 cate lamellfie disposed in the form of the letter V ; or again, as in what may be Trlco- 

 rythus (PI. XLI.). In the matter of proportional size, the tracheal branchia; of the first 

 abdominal segment (when they are developed at all) are frequently minute {e. g. in Ccenis 

 and Ephemera) ; the last pair is visually small ; the second pair is the largest in Ccenis, 

 the third or the fourth pair in many others ; but very often all of tliem are much alike 

 in size. The substance of tracheal branchiaj with the margin entire is stronger than 

 that of others whose form is less simple, being usually more or less coriaceous or 

 corneous. When they are unusually stout their hinder or under surface is lined with 

 delicate membrane, upon which the trachea? are distributed. Most of the tracheal 

 branchiae are kept in rapid agitation ; but the first and the last pairs, and in Ccenis the 

 elytroid pair, are usually held almost motionless. 



The branchial trachea? branch in a manner very similar to that of the ribs and veins of 

 dicotyledonous leaves of like form. They should be examined (when this is possible) 

 while the insect is yet alive, because very soon after death endosmosis drives the air out 

 of their ultimate subdivisions, which cannot then be seen. More than one skilled ana- 

 tomist, trusting to inspection of cabinet examples placed in preservative fluid, has 

 denied the existence of tracheae within filamcntose tracheal branchiae, being unaware of 

 the rapidity and completeness of their obliteration in the dead insect under ordinary 

 circumstances. 



Tracheal respiration is apparently carried on to some little extent by means of other 

 organs than the regular tracheal branchite in certain instances. The rectum, the 

 expanded borders of the head and front portion of the thorax in Ecclyurus and its allies, 

 the hinder surface of the femur in these and some other genera, and perhaps the caudal 

 setae, may be taken as examples of such accessory organs. 



