vol. 158, no. 4 the american naturalist october 2001 Notes and Comments On the Use of the Time Axis for Ecological Separation: Diel Rhythms as an Evolutionary Constraint Noga Kronfeld-Schor, 1,* Tamar Dayan, 1 Ralf Elvert, 2 Abraham Haim, 3 Nava Zisapel, 4 and Gerhard Heldmaier 2 1. Department of Zoology, Tel Aviv University, Tel Aviv 69978, Israel; 2. Department of Biology, Philipps University, D-35032 Marburg, Germany; 3. Department of Biology, University of Haifa-Oranim, Kiryat Tiv’on 36006, Israel; 4. Department of Neurobiochemistry, Tel Aviv University, Tel Aviv 69978, Israel Submitted November 13, 2000; Accepted May 18, 2001 Keywords: activity rhythm, body temperature rhythm, Acomys, temporal partitioning, evolutionary constraint, masking. Temporal partitioning as a mechanism of coexistence among competitors has seldom been explored and is con- sidered uncommon (Schoener 1986). Understanding eco- logical and evolutionary plasticity in use of the temporal niche axis is significant for assessing the potential for tem- poral partitioning and may be key to understanding the relative rarity of this phenomenon. Daily rhythms in mammalian physiology, hormone con- centrations, biochemistry, and behavior are driven by an internal circadian clock that is entrained to the predictable 24-h change in light intensity (Heldmaier et al. 1989; Re- finetti et al. 1992; Smith et al. 1998). Entrainment to this overriding cue (“zeitgeber”) allows mammals to choose the right time for a given response or activity without being easily misled by minor environmental disturbances (Aron- son et al. 1993). Thus, a degree of rigidity to environmental nonphotic stimuli is built into the circadian system. How- ever, under laboratory conditions, small mammals can be forced to shift their locomotor activity patterns by non- photic stimuli (Mistlberger 1991; Cambras et al. 1993; Challet et al. 1997; Brinkhof et al. 1998), indicating that activity is not tightly controlled by the circadian clock * E-mail: nogaks@post.tau.ac.il. Am. Nat. 2001. Vol. 158, pp. 451–457.  2001 by The University of Chicago. 0003-0147/2001/15804-0010$03.00. All rights reserved. (Mrosovsky 1999). As locomotor activity in mammals in- creases their body temperatures in proportion to the se- verity of activity (Minors et al. 1996), forced shifts of ac- tivity patterns also affect overt temperature rhythms. Whether a real shift in endogenous rhythmicity of activity and body temperatures is achieved with nonphotic stim- ulation, or whether the perceived shift is merely a masking effect, is still subject to debate (Edgar and Dement 1991; Edgar et al. 1991; Mrosovsky 1995; Meerlo and Daan 1998). Masking is defined as “any process that distorts the original output from the internal clock whether this orig- inates from inside or outside the body” (Minors and Wa- terhouse 1989, p. 30). As soon as the masking effect is removed, the underlying circadian rhythm is revealed (Wa- terhouse et al. 1996). Two ecologically similar spiny mouse species coexisting in rocky deserts of the Middle East (Shkolnik 1971; Kron- feld et al. 1996) allow an excellent test of the effect of competitively induced temporal partitioning on diel rhythms at the evolutionary scale and may provide insight into ecological aspects as well. The nocturnal common spiny mouse (Acomys cahirinus; mean body mas g) s p 34 and the diurnal golden spiny mouse (Acomys russatus; mean body mas g) overlap in their biology (Shargal s p 43 et al. 2000), diet (Kronfeld-Schor and Dayan 1999), and foraging microhabitat use (Jones et al. 2001). A key study demonstrated that when A. cahirinus is removed from a shared habitat, A. russatus shifts its activity time and be- comes nocturnal, implying that the two species compete (Shkolnik 1971). Generations of selection for diurnal activity in A. rus- satus have actually resulted in some adaptations, such as dark skin pigmentation and a high concentration of ascor- bic acid in its eyes (Koskelo et al. 1989). On the other hand, A. russatus retains the rod-based retina typical of nocturnal mammals (Kronfeld-Schor et al. 2001a), and the potential for nonshivering thermogenesis is similar in both species and appears evolutionarily constrained, al- though A. cahirinus is exposed to much colder tempera- tures in winter (Kronfeld-Schor et al. 2000) and spends