Carbon, nitrogen, and phosphorus stoichiometry of cyprinid fishes

Citation
Rw. Sterner et Nb. George, Carbon, nitrogen, and phosphorus stoichiometry of cyprinid fishes, ECOLOGY, 81(1), 2000, pp. 127-140
Citations number
54
Categorie Soggetti
Environment/Ecology
Journal title
ECOLOGY
ISSN journal
00129658 → ACNP
Volume
81
Issue
1
Year of publication
2000
Pages
127 - 140
Database
ISI
SICI code
0012-9658(200001)81:1<127:CNAPSO>2.0.ZU;2-O
Abstract
We investigated the carbon, nitrogen, and phosphorus levels in whole fish a nd gut samples of several species of cyprinids, relating our findings to nu trient flux models. Some differences in whole-fish nutrient content across species, lakes, and seasons, as well as differences across fish length and mass, were found. N and P contents were highest in fathead minnows and lowe st in pearl dace, with northern redbelly dace and finescale dace intermedia te. Larger fish had higher percent C and lower percent N and P. However, al l differences in whole fish C, N, and P chemistry were small. Cyprinids had the following mean composition: carbon, 46%; nitrogen, 9.7%; and phosphoru s, 1.5%. The cyprinid molar C:N:P ratio was 242:16:1. These values make cyp rinids relatively low in phosphorus compared to other fish that have been p reviously studied, especially members of the Percidae and Centrarchidae. Gu t contents were lower in N and P than the whole fish, and C:N and C:P ratio s were correspondingly higher in gut contents than in the whole fish. Thus, minnows must concentrate both of these nutrients within their biomass comp ared to what they eat. The N:P ratio of minnows and minnow gut contents had nearly identical means. All chemical variables showed lower variation in t he fish than in the gut contents, supporting a homeostatic model of nutrien t flux. Stable nitrogen isotope analysis found that minnows were similar to 3 parts per thousand (parts per thousand) heavier than their gut contents, providing evidence that gut contents analyzed were derived mainly from ing ested material. A homeostatic nutrient model appears to be an appropriate o ne for fish.