Evaporative water loss in scaleless snakes.

1. Rates of water loss were measured in two aberrant scaleless water snakes, Natriu sipedon, and in six normal animals. 2. Pulmocutaneous water loss of the scaleless animals was equal to or less than that of the controls at 20, 27, and 34°C. 3. The thermal dependence of pulmocutaneous water loss in all snakes was low (Q,, = 1.31-1.89). 4. The proportion of total water loss due to cutaneous evaporation (86.5%) in a scaleless animal at 20°C was similar to that previously reported for normal Ncrtris. 5. Thus, reptilian scales and their associated features (e.g. thick keratin layers, superficial dermal layer) cannot be considered adaptations for the curtailment of integumentary water loss. REPTILIAN skin is characterized by a low water permeability: cutaneous water loss in this group is low in comparison to all other vertebrates. However, the morphological and physiological bases of this low cutaneous permeability are poorly understood. The fact that a scaly integument is one of the most conspicuous features of the reptiles has led to the common hypothesis that scales may be important for retarding water loss. However, our recent observations (Licht & Bennett, 1972) on an abberant scaleless gopher snake, Pituoplzis catenfer, have cast doubts on this view. Despite gross morphological abnormalities in its integument, the rate of pulmocutaneous evaporative water loss in this scaleless animals was identical to that in a normal specimen. We concluded that reptilian scales and their associated features (eg. thickened beta-keratin and superficial dermis) cannot be considered adaptations to restrict water loss. Unfortunately, it is difficult to generalize on the basis of a single specimen. Further we did not partition water loss into pulmonary and cutaneous components; it is possible that the scaleless individual might have had a lower pulmonary water loss and hence a higher cutaneous loss than the control animals. However, such differences seemed unlikely since rates of oxygen consumption were the same in both animals. Also, water loss was measured only at a single body temperature and the contribution of scales to water retention may vary with temperature. The subsequent acquisition of two more scaleless snakes of another species, the water snake Natrix sipedon, members of a different subfamilial group and ecological type, allowed further examination of the importance of scales in water loss. In particular. we took this opportunity to examine the thermal effects and comp&tmentalization of water loss in relation to scale structure. * Present address: School of Biological Sciences, University of California, Irvine, CA 92664, U.S.A. IMATERIALS A N D ]METHODS Two scaleless specimens of the common water snake Narris sipedon were collected by David Roudebush in Harford County, Maryland, and were sent to us by Dr. George Zug of the U.S. National Museum. One of the specimens was a small juvenile (approximately 30cm and 9.6g) and the other (65cm and 82.6g) had been in captivity for several years. Both animals were similar in appearance; they possessed ventral scutes but completely lacked all dorsal and lateral scales. In this regard, they had even Sewer scales than the Pitirophis examined previously (Licht & Bennett, 1972). The body surface had a superiicially smooth texture and lacked the characteristic luster of the normal snake; their color patterns were essentially normal. For comparison, a series of three juvenile (7-1 1-5g) and three adult (84.8-125.8 g) Natrix were utilized a s control animals. For determination of total pulmocutaneous water loss, animals were placed individually in a darkened air-tight glass tube (1.0 m x 2.5 cm. diameter) containing a thermocouple to monitor temperature. Dry air was metered through this chamber a t a rate sufficient to maintain an internal water vapor pressure of 3-8 mm Hg (1@30% r.h.) The relative humidity of the excurrent air line was monitored with a Hygrodynamics narrow range humidity sensor connected to a Varian strip chart recorder. Water loss values were also checked gravimetrically by the collection of water vapor with Drierite (anhydrous CaSO,); values were identical to those obtained from the humidity sensor. The animal and sensing apparatus were placed in a controlled temperature cabinet set at 20, 27, o r 34°C (+0.5"C). The animal was left at each temperature for I hr to equilibrate and humidity readings were then recorded continuously for 2-3 hr. Animals remained quiet and inactive during measurement. The 15-min period during wrhich relative humidity readings were lowest was utilized to estimate minimum rates of water loss. The humidity readings never varied more than 1% during the minimum measurement period. Attempts to partition pulmonary and cutaneous water loss were generally unsuccessful because the skin of the scaleless animals was extremely delicate and tore easily. However, one successful experiment in which the head was isolated from the body by use of a rubber membrane allowed estimation of the cutaneous component at 20°C in the smaller scaleless individual. Table 1. Minimal rates of pulmocutaneous water loss in scaleless and normal N a e i x . Water loss is reported as mg H,O per g body wt/hr Wt. Temperature Specimen g 20°C 27°C 34°C Small Scaleless Normal Normal Normal Large Scaleless Normal Normal Normal