Bl. Peckarsky et al., Hydrologic and behavioral constraints on oviposition of stream insects: implications for adult dispersal, OECOLOGIA, 125(2), 2000, pp. 186-200
The supply of recruits plays an important role in plant and animal populati
on dynamics, and may be governed by environmental and behavioral constraint
s on animals. Mated females of the mayfly genus Baetis alight on rocks prot
ruding from streams, crawl under water and deposit a single egg mass under
a rock. We surveyed oviposition and emergence of a bivoltine population of
B. bicaudcatus in multiple stream reaches in one high-altitude watershed in
western Colorado over 3 years to establish qualitative patterns at a regio
nal scale (entire watershed), and quantitative patterns over six generation
s at a local scale tone stream reach). We also measured characteristics of
preferred oviposition substrates, performed experiments to test hypotheses
about cues used by females to select oviposition sites, and measured mortal
ity of egg masses in the field. Our goals were to determine whether: (1) hy
drologic variation necessitated dispersal of females to find suitable ovipo
sition sites; (2) the local supply of females could provide the supply of l
ocal recruits; and (3) local recruitment determined the local production of
adults. The onset of oviposition corresponded with the decline of spring r
un-off, which differed dramatically among years and among sites within year
s. However, eggs appeared before any adults had emerged in 8 of 22 site-yea
rs, and adults emerged 2-3 weeks before any eggs were oviposited in 3 site-
years. Furthermore, the size distribution of egg masses differed from that
predicted by the size distribution of females that emerged from seven of ni
ne stream reaches. Protruding rocks and eggs appeared earlier each summer i
n smaller tributaries than in larger mainstream reaches, suggesting that hy
drologic and behavioral constraints on oviposition may force females to dis
perse away from their natal reach to oviposit, and possibly explain the pre
dominantly upstream flight of Baetis females reported in other studies. Loc
al oviposition rates in one third-order stream-reach increased rapidly as s
oon as substrates protruded from the water surface, and females preferred l
arge rocks that became available early in the flight season. However, femal
es oviposited on <10% of all available rocks, and <65% of preferred rocks a
s determined by an empirical model. These data indicated that the timing of
appearance of suitable oviposition sites determined the phenology of local
recruitment, but that preferred oviposition sites were not saturated. Thus
, the magnitude of local recruitment was not limited by the absolute abunda
nce of preferred oviposition sites. Only 22% of egg masses observed in the
field suffered mortality during their embryonic development, and per capita
Baetis egg mass mortality was significantly lower on rocks with higher den
sities of egg masses. Thus, we suspect that specialized oviposition behavio
r may reduce the probability of egg mortality, potentially compensating for
the costs of dispersal necessary to locate suitable oviposition sites. Fin
ally, the number of adults that emerged at one stream reach was independent
of the number of egg masses oviposited over six generations of Baetis; and
local recruitment was not a function of the number of adults of the previo
us generation that emerged locally. The patterns of oviposition and emergen
ce of Baetis found in this study are consistent with the following hypothes
es. Recruitment of eggs in a stream reach is not limited by the local suppl
y of adults, but is a function of the regional supply of dispersing adults,
which are constrained by the spatial and temporal distribution of preferre
d oviposition habitat.
Furthermore, subsequent local production of adults is not a function of the
supply of recruits, arguing for post-recruitment control of local populati
ons by processes operating in the larval stage (e.g., predation, competitio
n, dispersal, disturbance). Processes affecting larval and adult stages of
Baetis act independently and at different scales, thereby decoupling local
population dynamics of successive generations.