Deductive and Inductive Methods of Accumulating Reliable Knowledge in Wildlife Science

Abstract: My goal was to compare deductive and inductive methods of accumulating reliable knowledge in wildlife science. Under the hypotheticodeductive (H-D) method, observations are used to formulate explanatory or causal hypotheses, which serve as the basis for deductions (predictions) of expected events. Field experiments are designed to determine whether the deductions hold, in which case hypotheses are tentatively accepted or otherwise rejected. The H-D method provides the only way to test research hypotheses, but in field ecology it can lead to ambiguity and error. The method: 1) does not preclude confusion of correlation and cause, 2) might perform deceptively in multiple-cause venues, 3) is algorithmically blind to the fact that different hypotheses can lead to the same deduction, and 4) lacks an impartial means of determining whether a deduction has been observed and, therefore, whether a hypothesis is meritorious. Under the process of induction, the results of a study are presumed to hold generally and taken as knowledge accordingly. Induction is much maligned by logicians and philosophers, and wildlife scientists have built false knowledge inductively. However, wildlife scientists have auxiliary knowledge such as facts of natural history to screen inductions for validity. Both the H-D method and induction have important roles in the accumulation of reliable knowledge in wildlife science.     

DISPERSAL BY YEARLING MALE WHITE-TAILED DEER AND IMPLICATIONS FOR MANAGEMENT

Abstract: The scale at which populations use the landscape influences ecological processes and management decisions. Dispersal and home-range size define the scale of landscape use for many large-mammal species. We measured dispersal and home-range size of yearling male white-tailed deer (Odocoileus virginianus) in southern Texas, and compared our results to values from the literature to understand the implications of dispersal in management of deer populations. We used radiotelemetry to monitor 22 yearling deer on 1 study site from October 1998 to October 1999, and 27 yearling deer on a second study site from October 1999 to October 2000. On the 2 study sites, 68% and 44% of yearling deer established new areas of use 4.4 ± 1.0 km and 8.2 ± 4.3 km, respectively, from the center of their autumn home range. Yearling males with spike antlers (2 points) were less likely to disperse than yearlings with fork antlers (>2 points) on 1 study site. Computer simulation showed that the scale at which deer use the landscape is large compared to property sizes in southern Texas and probably in other areas of the white-tailed deer's range. Differences in scale between land ownership patterns and landscape use by deer may result in a failure to meet management objectives and conflict among managers. High harvest rates for male deer occur in part because deer movements are large relative to property size, creating a “tragedy of the commons.” Cooperative management groups are beneficial if all landowners in an area agree on management objectives. Otherwise, deer-proof fences often are erected to reduce conflicts among property owners.     

Heterozygosity for recessive lethals in homosporous fern populations: Thelypteris palustris and Onoclea sensibilis

Sporophytes from natural populations of Thelypteris palustris and Onoclea sensibilis were subjected to breeding tests to reveal the presence of recessive lethal alleles in their genotypes. All sporophytes were found to be heterozygous for recessive sporophytic lethals. Despite the capability for producing completely homozygous genotypes by intragametophytic self-fertilization, ferns in natural populations seem to retain considerable genetic loads of deleterious recessive alleles. It is suggested that natural selection operates to strongly favour heterozygous rather than homozygous sporophytes under most conditions.     

The Quilon Limestone,Kerala Basin,India: an archive for Miocene Indo‐Pacific seagrass beds

Reuter, M., Piller, W.E., Harzhauser, M., Kroh, A., Rögl, F. & ?ori?, S. 2010: The Quilon Limestone, Kerala Basin, India: an archive for Miocene Indo‐Pacific seagrass beds. Lethaia, Vol. 44, pp. 76–86. The facies of the fossiliferous Quilon Limestone in SW India is described for the first time in detail at the Padappakkara‐type locality. Facies (fossiliferous, micrite‐rich, bioturbated sediment with intercalated sand pockets) and faunal composition (epiphytic foraminifers, seagrass feeding Smaragdia gastropods, bioimmuration of celleporiform bryozoan colonies) indicate a seagrass environment. The large discoidal archaiasin foraminifer Pseudotaberina malabarica, in particular, is considered as a proxy for seagrass communities. Recent seagrasses have their centre of generic richness in the Indo‐Pacific where they cover wide areas in the tidal and shallow sub‐tidal zones. However, their geological record is only fragmentary and their palaeobiogeographic distribution has a big stratigraphical gap in the Miocene Western Indo‐Pacific region. The described nannoplankton flora and planktonic foraminifers from the Quilon Formation demonstrate that the deposition of the studied seagrass bed occurred in nannoplankton biozone NN3. This timing suggests formation during the closure of the Tethyan Seaway. The Quilon Limestone is thus an early Western Indo‐Pacific seagrass bed and an important step in reconstructing the history of seagrass communities. □Quilon Formation, Pseudotaberina malabarica, seagrass facies, Burdigalian, Indo‐Pacific.     

CHORUS HOWLING BY WOLVES: ACOUSTIC STRUCTURE,PACK SIZE AND THE BEAU GESTE EFFECT

ABSTRACT

A variety of structural parameters were measured from wolf choruses recorded in the Superior National Forest, Minnesota, USA. Mean duration of 60s did not vary with pack size or composition. Packs replied to simulated howling after an average of 40s, often interrupting the stimulus howls. Choruses began with simply-structured howls, which became increasingly modulated as the chorus progressed. Little difference in mean fundamental frequency or other howl parameters was found among the choruses from packs of various sizes and compositions. In particular, choruses produced by single adult pairs did not differ from those of larger packs accompanied by pups. The lack of relationship between chorus parameters and pack size or composition indicates there is little useful information concerning a pack's size to be found in its chorus howling.

The observation that chorus howling by adult pairs is often perceived as that of larger groups with pups suggests that chorus structure has evolved to exaggerate the apparent size of the pack, especially those newly-established or otherwise reduced in number. If so, wolf howling choruses may represent a mammalian example of the Beau Geste effect, made particularly viable because of the relative immunity of the signal to probing.     

Quantitation of malaria sporozoites in the salivary glands of wild Afrotropical Anopheles

The number of malaria sporozoites in the salivary glands was determined microscopically for 1137 wild, naturally infected Anopheles from western Kenya. Infective Anopheles gambiae Giles sensu lato (n = 874) contained a geometric mean (GM) of 962 sporozoites and An.funestus Giles (n = 263) contained 812. No significant differences were detected in geometric mean numbers of sporozoites between species, collection techniques or sites. Of the infective An.gambiae, 1.7% (15/874) contained more than 41,830 sporozoites, the maximum observed for An.funestus. Microscopic techniques were found to be more sensitive than enzyme-linked immunosorbent assays (ELISA) for detecting low-grade sporozoite infections in salivary glands. Salivary gland sporozoites from 83.6% of the 1137 gland infections were identified by ELISA as either Plasmodium falciparum Welch (n = 910), P.ovale Stephens (n = 7), P.malariae Grassi & Feletti (n = 3) or mixed (n = 30). The 187 gland infections which could not be identified by ELISA contained significantly fewer sporozoites (GM = 242) than those which could be identified (GM = 1200).