Purpura papulosa (lichen lividus, Willan) is a form of purpura where, in the midst of ecchymoses, livid papules appear. These probably depend upon a large amount of hemorrhage occurring within a limited space, most often surrounding the orifices of hair-follicles, because these are supplied with a capillary network that comes directly from the deeper layer.[8] They are formed most abundantly on the legs of scrofulous, cachectic persons who have purpura. Care must be taken to distinguish this form of purpura from erythema multiforme and erythema nodosum, where blood is usually extravasated secondarily into the tissues. Those cases only where the purpura is primary should be recognized as purpura papulosa. [Footnote 8: Hebra, _Skin Diseases_, New Syd. Soc. Transact., ii. p. 425.] The purpuric effusion appears to act as an irritant upon the tissues, and to excite inflammation. Gangrene of the mucous coat of the intestines has resulted from extensive hemorrhagic extravasations, and from a similar cause cutaneous gangrene has been known. These complications, however, are rare. ETIOLOGY.--The immediate causes of purpura are quite unknown. Both sexes and persons of every age are affected by it. While it is most often seen in debilitated subjects, those in vigorous health possess no immunity. It has often been observed during convalescence from other maladies. It cannot be said that those who are miserably clothed, fed, and lodged are especially predisposed to attacks of purpura. Between purpura and hæmophilia, etiologically, there are many points of difference. Purpura is not hereditary, nor is there a purpuric diathesis in the strict sense of the term. Some persons, indeed, seem to possess a {191} predisposition to the disease, and some authors claim for purpura rheumatica a distinct annual type. This, however, is not at all certain. Recently it has been claimed that purpura hæmorrhagica depends upon the presence of a minute organism in the blood. Petrone[9] injected blood drawn from patients with this disease under the skin of rabbits, producing widely-distributed hemorrhages. In the blood of these individuals and of the injected rabbits micrococci and bacilli were detected. Watson Cheyne[10] also describes a plugging of the capillaries with bacilli. These were 1/7700 of an inch in length and 1/20000 of an inch in diameter, and were arranged in colonies. In another case there were found micrococci arranged in chains. These swarmed in the capillaries and some larger vessels, and sometimes completely blocked them. Although an origin in infection has thus been claimed for purpura hæmorrhagica, the fact that more than one variety of micro-organism was observed cannot fail to excite suspicion of, possibly, erroneous observation. [Footnote 9: _Lo Sperimentale_, 51, 1883.] [Footnote 10: _Lancet_, i., 1884, 344.]