234 EEV. A. E. EATON ON RECENT EPIIEMERID.E OR MAYFLIES. 



Li'oacl at the base of the wing, becomes linear beyond the elbow, and the axillar region is 

 narrow. In Ble])tHs (PI. LXV. 1) the hind wing is abnormal. The 6 oculi are almost 

 contiguous with one another above ; and, in ? , the median ocellus is rather smaller tlian 

 the hinder ocelli. 



The items chiefly relied upon for generical classification in this Section are differences 

 noticeable in the relative leugtlis of joints of the tarsi, and in the forms of the penis-lobes 

 of adult specimens ; and also peculiarities of the nymphs. The style of coloration of 

 the wings of subimagines, and the femoral markings (if any) of the flies are likewise of 

 some critical value. But to arrive at a true decision respecting the genus of an individual 

 representative of this type upon these bases, caution and extreme circumspection are 

 often demanded, in view of the following circumstances. 



The relative lengths of the tarsal joints of corresponding legs are not strictly invariable 

 in every species of a genus, nor in all specimens of the same sex of a species. The 

 variations noticeable in the fore tarsi of a large series of examples of one species, like the 

 differences between the tarsi of some species and of other species of one genvis, may 

 reasonably be suspected to be due either to circumstances attending the last moult, or 

 (in some instances) to marked diversity in the nature of the habitats of the individual 

 nymphs. But apart from such variations, it sliould always be remembered that tarsi are 

 very subject to deformity. The deformity of a leg in course of reproduction is conspicuous, 

 and familiar to physiologists ; no entomologist w^ould be likely to mistake such a leg for 

 a well-proportioned limb. But it does not appear to have been noticed hitherto, that 

 when a nymph has chanced to sustain the loss of a limb in early life, the remaining leg 

 of that pair, although uninjured, is apt to deviate from its proper projiortions in the 

 adult fly, through some of its component parts becoming hypertroj)hied. In this way 

 one or more of tlio first two or three joints of either of the hinder tarsi may acquire 

 abnormal extension in so slight a degree as to be appreciable only by specialists, and 

 yet quite sufficient in amount to occasion perplexity should unblemished specimens be 

 unobtainable. 



When Plate XXIV. was published, this characteristic defect in adult flies grown up 

 from maimed nymphs had not been discovered. Some of the legs there figured were 

 undoubtedly the fellows of limbs in course of reproduction. I have therefore re-drawn 

 and re-measured the legs of representative species of all the genera of this Section 

 (excepting Pcegniocles and Cumpsoneuria), and quote the results thus obtained both in 

 the tabulated admeasurements subjoined, and in the descrijitive lettei'press, in lieu of the 

 older data. 



The aspect of the petiis in dead specimens is sometimes much altered from that 

 normally presented by it during life by convulsive movements made by the moribund 

 insect. Sometimes these lead to extreme extrusion of the lobes accompanied by excessive 

 lateral divergence ; but sometimes the contrary action is set up, and the lobes are 

 withdrawn by a process of intussusception either partially or completely, leaving in the 

 latter case nothing but the stimuli visible. 



In the nymphs generical distinctions reside chiefly in the abdominal tracheal-branchiye, 

 in some of the mouth-parts, and in the structure of the pronotum. In re^^ose their 



