Body size predicts ontogenetic nitrogen stable-isotope (δ15N) variation, but has little relationship with trophic level in ectotherm vertebrate predators.

Large predators have disproportionate effects on their underlying food webs. Thus, appropriately assigning trophic positions has important conservation implications both for the predators themselves and for their prey. Large-bodied predators are often referred to as apex predators, implying that they are many trophic levels above primary producers. However, theoretical considerations predict both higher and lower trophic position with increasing body size. Nitrogen stable isotope values (δ15N) are increasingly replacing stomach contents or behavioral observations to assess trophic position and it is often assumed that ontogenetic dietary shifts result in higher trophic positions. Intraspecific studies based on δ15N values found a positive relationship between size and inferred trophic position. Here, we use datasets of predatory vertebrate ectotherms (crocodilians, turtles, lizards and fishes) to show that, although there are positive intraspecific relationships between size and δ15N values, relationships between stomach-content-based trophic level (TPdiet) and size are undetectable or negative. As there is usually no single value for 15N trophic discrimination factor (TDF) applicable to a predator species or its prey, estimates of trophic position based on δ15N in ectotherm vertebrates with large size ranges, may be inaccurate and biased. We urge a reconsideration of the sole use of δ15N values to assess trophic position and encourage the combined use of isotopes and stomach contents to assess diet and trophic level.

Haemogregarina spp. in a wild population from Podocnemis unifilis Troschel, 1848 in the Brazilian Amazonia.

The study objectives were to characterize the morphology of the parasitic forms and describe the prevalence and intensity of Haemogregarina spp. in a population of the turtle Podocnemis unifilis as well as to examine the relationships between parasitism and turtle variables such as gender, size, and weight. Samples were taken in the Tapajós and Jamanxim rivers, Itaituba, Pará state, Brazil. Blood was collected from the tail vein of 72 P. unifilis specimens, including 35 males, 36 females, and one unsexed juvenile. The prevalence of Haemogregarina spp. was 98% (n = 71). The mean parasite intensity of Haemogregarina spp. was 118 (1-582) parasites/2000 blood cells (6%). There was no significant difference in the mean parasite intensity between male (137.68 ± 121.8, n = 35) and female turtles (101.42 ± 123.59, n = 35). There was no relationship between parasite intensity and carapace length. Although the relationship between parasite intensity and host body weight was significant, the relationship was weak. This is the first study on Haemogregarina parasitism with a relatively high number of turtles in Brazil.