Krasnov et al. 2008

Krasnov BR, Shenbrot GI, Khokhlova IS, Mouillot D, Poulin R. 2008. Latitudinal gradients in niche breadth: empirical evidence from haematophagous ectoparasites. Journal of Biogeography 35: 592-601.

This paper examined Rapoport’s rule in fleas that use small mammals in the Palaearctic as hosts, applying phylogenetic independent contrasts (PIC) to questions of geographic range size, latitude, and niche breadth. These authors consider Rapoport’s rule, of increased geographic range sizes with higher latitudes, to be a special case of a more general pattern of niche-breadth expansion (less specialization) with higher latitudes. Under this explanation, specialized species have narrow tolerances of abiotic conditions, use a small range of resources, and / or tolerant of a very limited set of competitors, predators, parasites, and diseases.

There is controversy in the current biogeographic literature about the extent of application of Rapoport’s rule or Rapoport’s effects. Some authors consider it a global phenomenon, driven by global mechanisms such orbital dynamics (Dynesius & Jansson, 2000) or habitat stability (MacArthur 1955; 1972) differences across the globe. Other authors, most notably Rhode (1996; 1999) consider Rapoport’s effects to be localized to only some latitudes or taxonomic groups.

Recently, Vazquez and Stevens (2004) presented a hypothesis for a mechanism underlying a global Rapoport’s rule. Briefly, they proposed that a positive relationship between niche breadth and latitude will occur if 1) there is a latitudinal gradient in species richness and 2) the species interaction network is an asymmetrically specialized interaction network such that specialists tend to interact with generalists. This hypotheses appears to involve the opposite direction of causality compared to hypotheses relating high tropical species richness to narrow species niches (i.e. high specialization) via character displacement and competitive exclusion.

The measure of niche breadth used in this study includes an estimate of the taxonomic distinctiveness of the fleas’ hosts. A flea species with a broad niche will use hosts that are more distantly related to each other than will a flea species with a narrow niche, even if both flea species use the same number of species of host. The authors describe this as the STD index, in which high values indicate more taxonomically distinct hosts such as hosts in different orders. In their analysis, these authors excluded all extreme specialist flea species, those found in only one geographic site or on only one mammal host species.

Analyses included regressions of two dependent variables (host specificity and geographic range size) against the independent variable (geographic range position) using both conventional statistics and PIC. Geographic range position was taken as the midpoint latitude of a species geographic range, which was computed by constructing minimum surface polygons and other techniques more fully described in their methods section.
Species level phylogenies were constructed for each family of fleas, based on a family-level phylogeny previously produced by Medvedev (1995) and morphological and taxonomic characters. Polytomies were considered “soft” i.e. they assumed no knowledge of hidden branching patterns. The PIC conducted by these authors included a range of sophisticated statistical controls, again more fully detailed in their methods section.

The results of this study demonstrate 1. fleas follow Rapoport’s rule, at least in the Palaearctic and 2. host specificity in fleas declines at higher latitudes. Thus the positive relationship in fleas of niche breadth and latitude holds for both abiotic (geographic range size) and biotic (diversity of hosts) components of their niches. There were some exceptions, but the overall trend is clear. These authors propose a mechanism underlying this trend that is much in line with Vazquez and Stevens (2004), involving interactions between niche breadth and geographic range and latitude and niche breadth.