This paper comments on tree-climbing methods presented by Perry (1978) and gives alternative techniques, safety hints, and sources for more information and equipment used in rope and descending. DONALD PERRY'S (1978) PAPER describing a method for climbing tall rain forest trees is welcome in that it introduces tropical biologists to specialized techniques known mostly to cavers and rockclimbers. A potential hazard exists, however, if persons lacking experience with such techniques are to learn them from Perry's paper and then apply them without further guidance. The purpose of this paper is to point out one specific lapse of safety precautions in Perry's paper, to give several tips on the safe use of rope-ascending techniques, and to point out a few reliable sources of equipment and information on such techniques. The techniques described here have been used by the author during 10 years of vertical caving and rockclimbing, and in two seasons of climbing tropical forest trees in the course of a study of the bat falcon (Falco rufigalaris). Figure 1 of Perry's paper shows non-locking carabiners (steel or aluminum snap-links) linking the jumars (mechanical ascenders) to the slings which support the climber. This practice should never be permitted, as non-locking carabiners can be accidentally opened by contact with rope, limbs, etc., resulting in a fall. Furthermore, the use of evenlocking carabiners should be avoided whenever it is possible to eliminate them by affixing slings directly to the ascenders by means of a girth hitch (fig. 2d) or other suitable knot. The philosophy of anyone using climbing equipment should be to minimize the number of links in the system, since each link is a potential site for equipment or operator failure. As a rule, any carabiner linking the climber to his/her or rappeling gear should be of the locking variety. Steel-locking carabiners are preferable to aluminum ones for this application, as the latter may experience damage to threads as a result of the harness sling hanging up on the collar of the locking mechanism. This problem can also be elimi1Present address: Department of Zoology, University of California, Davis, California 95616, U.S.A. nated by attaching the sling to the carabiner by means of a girth hitch (fig. 2d) which prevents back-and-forth movement of the sling which can result in the hanging-up referred to. If locking carabiners are not available, two regular carabiners should be used in combination, with the gates on opposite sides. Another important safeguard is to have more than one ascender linked to the climber's harness so that the failure of one ascender or sling will not result in a fall. This may be achieved in Perry's system by adding a loop of webbing or rope connecting the lower (feet) jumar to the parachute harness. This loop should be long enough to be slack during normal operations but will prevent a fall should the upper ascender or its sling fail. While this type of ascender has been known to disengage inadvertently from the climbing rope, primarily when the climber is ascending along a horizontal or diagonal rope, it is, nonetheless, probably the safest and most versatile mechanical ascender available and is enjoyable to use with the proper precautions. Though a person may prefer to use mechanical ascenders such as the jumars pictured by Perry, anyone using such ascenders should also be aware of the method for tying and using prusik knots. These knots can be made of manila, nylon, or polypropylene ropes which are relatively cheap and available in the most remote villages. Thus, this system has the advantages of economy and utility in unanticipated ascents or emergencies. If frequently inspected for wear and replaced judiciously, prusik knots are probably safer than mechanical ascenders; many cavers use them as their sole means of ascent. Prusik knots work as slip-knots; they can be moved up or down the rope after a quick loosening motion, but catch and hold when weight is applied. Though some systems employ only two knots, the three-knot system is inherently safer as it provides an additional 286 BIOTROPICA 13(4): 286-291 1981 This content downloaded from 157.55.39.153 on Mon, 19 Sep 2016 04:39:18 UTC All use subject to http://about.jstor.org/terms