INTESTINAL KESPIRATION JN ANNELIDS. 785 



alimentary canal, whose chief function is to store up the oxygen secreted by the 

 intestine, in order that it may be used at need, alone or mixed with sea-water, for 

 vicarious respiration. They are. however, of necessity also hydrostatic. 



In the Syllids, the bladders are best developed in what are apparently the more 

 primitive forms ; in the more modern genera they seem to be disappearing, perhaps 

 because their respiratory function is being replaced by that of the general surface of 

 the body. The peculiar appendages of the alimentary canal in Nereis are probably 

 homologous. Finally Phi/llodoce, though possessing no bladders, may be observed at 

 times to have a large quantity of gas in its intestine. It is to be noted that Hesionids, 

 Syllids, Phyllodocids, and Nereids have no gills. 



In conclusion, " alimentary respiration constitutes a function which belongs in 

 greater or less degree to the whole group of Annelids, and indeed represents, along 

 with cutaneous respiration, the most primitive form of respiratory activity. We may 

 accordingly suppose that, as a preponderating cutaneous respiration has led to the 

 development of external appendages in the form of gills, so a preponderating alimentary 

 respiration has led to the development of bladders." 



Observations on the respiratory vesicles of Syllids were subsequently published by 

 DE St Joseph (14), who controverts some of Eisig's conclusions. This investigator also 

 often saw Syllids swallow both air and water (this is definitely stated for Syllis 

 alternosetosa and »S'. prolifera) ; other specimens were seen whose intestine contained 

 both air and water at the time of examination ; and in others again air was seen to be 

 expelled by mouth or anus. These observations were made in many cases on forms 

 which have no " air-bladders ' ; hence the ingestion of air and water does not depend 

 on the presence of these organs. 



The bladders are not "swim-bladders" {i.e. are not hydrostatic in function), since 

 when expanded they are filled with water, never with air. The hydrostatic function 

 is performed by the intestine ; thus the Syllids which were oftenest observed floating 

 (species of Tiijpanosyllix and EurysylUs) have no bladders, but show constrictions of 

 the intestine, with intervening pockets in which the air is contained. The author 

 considers the true seat of production and reservoir of the gas to be the intestine, which 

 extracts the air from the sea-water as EisiG thinks (illisici's idea is, however, rather that 

 of a definite secretion of the gas by the intestinal epithelium), or into which the air 

 penetrates directly, by the mouth. 



In a subsequent instalment of the paper (1.5) similar observations on other tamilies 

 are described. Leptonereis vaillantl (Lycoridiu) sometimes puts forth its proboscis in 

 order to take in air and water, which are received into two extensible respiratory 

 appendages of the stomach. The Phyllodocids also aspirate air and water through 

 their extended proboscis. In Magalia 2ye7'a7'mata (Hesionidoe) there are no swim- 

 bladders ; the distensible ventriculus and the intestine, however, may contain air and 

 water introduced through the proboscis. O.rydronius propi7iquiis often projects its 

 proboscis, swallowing air, water, and small particles within its range ; these are then 



