GENERAL FEATURES OF WATER EXCHANGES 



183 



body size so closely that individuals of all species drinking side by 

 side, after depletion of water, would on the average gulp simul- 

 taneously and take drafts of equal duration. 



The quantities concerning water intake might profitably be 

 observed in other classes of animals for which information now 

 scarcely exists. Frequency of drafts is one such; the snail Helix 

 is known to ingest only once or twice a week (Howes and Wells, 

 '34) ; certain hummingbirds drink every few minutes while on the 

 wing (Ditmars, '37, p. 131), butterflies and other insects take water 

 at intervals quite unknown (Buxton, '32b). Still others like the 

 frog and the termite Termopsis die, presumably of water deficit, 

 without attempting to ingest by mouth water that is available or 



06 

 Hours 



Fig. 105. Course of relative water load (% of initial load) after recovery from 

 deficit began in diverse species of invertebrate animals. Zoothamnhim put from 0.05 M 

 sucrose into fresh water; initial AW -22; body volume 1.3x10"* ml.; data of Kitching 

 ('38, p. 147). Helix pomatia after 4 months' privation of water; initial AW -33; 

 body weight 14 gm. ; water placed on mantle only and not on mouth; data of Kiinkel 

 ('16, p. 45). Phascolosoma put from 150 per cent sea water into normal sea water; 

 initial AW -26; average for 5 individuals, B; body weight 1.52 gm. each; 22° C. ; further 

 data of Adolph ('36b). Bipalium after 10 hours' desiccation; initial AW -40; body 

 weight 1.0 gm.; about 30° C; data of Kawaguti ('32, p. 52), 



at least present (Cook and Scott, '32). Very numerous are the 

 possible functions with which frequency of drinking is correlated, 

 when water is continuously available. Numerous are the tales 

 extant about habits of drinking in various species ; they need not 

 be considered further in this investigation. 



Time relations during recoveries in several sorts of living units 

 may be expressed in whole series of tolerance curves upon common 

 scales of load and time (fig. 106). The range of loads studied is 

 very diverse, making for moderate diversity in the magnitudes of 

 exchanges. 



