REU: Kallin Tea
A geographic comparison of willow leaf beetles (Chrysomela aeneicollis) and its predation by flies (Parasyrphus melanderi)




The effect of high elevation on organisms is a field that has been extensively studied. As the elevation increases, the proportion of oxygen in the air becomes lower and the temperature generally decreases. These factors are known to be key components to what regulates an organism’s metabolism and reproduction, but a great deal is still unknown.
In this study, an elevation gradient comparison across three drainages demonstrates that Chrysomela aeneicollis does not react to elevation differences by changing the size of the egg clutches it lays. Concurrently, the survivability of its offspring does not seem to be effected by elevation differences.
In nature, C. aeneicollis has a specialist fly predator, Parasyrphus melanderi. It lays its eggs amongst the C. aeneicollis egg clutches and the hatchlings attack both eggs and larvae of C. aeneicollis. The survival rates of C. aeneicollis egg clutches were significantly lower when P. melanderi eggs were present in the clutch. However, no difference in predation was seen correlating to elevation differences. One surprising finding was that although P. melanderi had been a very strong predator at Big Pine Creek in the past, it was not present at this drainage this season. Also, counter-intuitively, the survival rates of C. aeneicollis egg clutches were not different from the other drainages, despite the absence of one of its primary predators.


One of the most challenging aspects of living in high altitudes for humans is the lower partial pressure of oxygen in the air. This often leads to respiratory complications that, in the long term, may lead to other health compromises relating to reproduction and metabolic processes in general. These high altitude effects have been shown to effect mammals, but it has not been studied in the willow leaf beetle, Chrysomela aeneicollis. In this study, we explore to see if this specie were susceptible to such difficulties at high elevations by monitoring the size of its egg clutches and its rate of survival along an elevation gradient.
C. aeneicollis is a beetle that feeds only on willow trees, mostly of the species Salix Orestera. In California, it seems to be most abundant in cool and moist habitats in coastal and riparian corridors (Rank and Smiley, 1994). In the Eastern Sierra Nevada Mountain Ranges, C. aeneicollis is often found between the elevations of 2500-3300 meters (Rank and Smiley, 1994). They typically do not go any higher most likely due to the fact that willow trees do not grow higher than tree-line, and this is the only plant on which they feed. As for their lower elevation limit, it may be due to stronger predation by ants lower down. The warmer temperature most likely is not the reason because when they are brought to a controlled laboratory setting and put in a warmer temperature environment, they do fine. However, C. aeneicollis has been known to fly away from a willow grove in search of better habitat. According to Dr. John Smiley, one possible impetus for such a move may be particularly hot temperatures that drove them to find somewhere moist and cooler. He also hypothesizes that since they often congregate near waterfalls and fast moving streams, perchance they have some way of detecting these geographic formations.
Another factor that may indicate high elevation differences is the rate of predation of C. aeneicollis by the fly Parasyrphus melanderi. This particular fly is a specialist that feeds only on C. aeneicollis to survive. The female P. melanderi would lay its eggs right between the eggs in the C. aeneicollis egg clutches. Then, when the hatchlings come out, they eat both the C. aeneicollis eggs and larvae. This fly can often be found wherever C. aeneicollis are found.
For this study, we chose to use three drainages in the Eastern Sierra Nevada mountains: Rock Creek, the south fork of Bishop Creek, and Big Pine Creek. (Add coordinates later) Within each drainage, three different study sites were chosen along an elevation gradient that would be comparable to each other. In Rock Creek, we used Heart Lake (3190m), Mosquito Flat (3111m), and Pine Grove (2837m). In Bishop Creek, the study sites were Pipeline (3131m), South Lake Parking Lot (2996m), and La Hupp (2854m). Lastly, the highest site at Big Pine Creek was the Upper Site (3229m), then Pond Bog (3128m), and Falls Site (2997m). All of these locations have been previously surveyed for adult C. aeneicollis presence, so we justly expected to find egg clutches to monitor later in the season.

Materials and Methods

In mid July 2004, a total of 170 C. aeneicollis egg clutches were marked with small aluminum plant tags (Table 1). Some of the study sites were not used after all due to difficulties with finding egg clutches. These sites consist of: Heart Lake and Pine Grove at the Rock Creek drainage, and Falls site at the Big Pine Creek draiange. All of the other study sites had 30 C. aeneicollis egg clutches marked, with the exception of Mosquito Flat, which had 20 clutches marked.
The number of C. aeneicollis eggs per clutch were counted, and any presence of P. melanderi eggs were also noted. Within the next 10 days, each egg clutch was re-visited, and a follow-up count of surviving eggs was performed. In some instances, some of the C. aeneicollis eggs had already hatched. In these cases, the number of larvae that were alive were counted, and the stage of development was also recorded. However, when calculating the percentage of survival, no distinction between egg or larval stage was made. If it were either an intact unhatched egg or hatched larva, it was counted as having survived. The eggs that were counted as dead were visibly destroyed by either looking extremely flaccid or completely trampled over.
The statistical program JMP 4.0 was used for all of the statistical tests conducted in this study. The sample size for each study site was 16 C. aeneicollis egg clutches. Only C. aeneicollis egg clutches that were of good quality were used in the statistical tests. The egg clutches that were excluded either had discoloration, or some egg damage that made the number of viable eggs difficult to determine.
The first test that were run was nested ANOVA comparisons across all three drainages to C. aeneicollis egg clutch size. A non-parametric Wilcoxon test was used to analyze mean survival percentages across drainages. ANOVA was also used to see if there was a correlation between C. aeneicollis egg clutch size and the number of P. melanderi eggs laid in them. A Pearson’s test was used to distinguish difference in P. melanderi egg presence in the different drainages, and a non-parametric Wilcoxon test was used to analyze the correlation between P. melanderi egg presence to C. aeneicollis egg survival in the South Lake Drainage. A non-parametric Spearman’s Rho test was used to explore correlation between elevation and multiple factors: C. aeneicollis egg clutch size, P. melanderi egg numbers, C. aeneicollis adult abundance earlier in the season, and C. aeneicollis larval abundance closer to the end of the season.


At Big Pine Creek, 0 out of 32 C. aeneicollis egg clutches had any P. melanderi eggs. This was statistically different from South Lake and Rock Creek (Fig. 1, P < 0.0001). However, Big Pine Creek C. aeneicollis egg clutch survival was not higher than the other drainages. In fact, across all three drainages, there was no statistical difference (Wilcoxon, P = 0.368). In 1988, Nathan’s paper demonstrates that P. melanderi had a strong presence in the past. No differences seen across drainages in C. aeneicollis egg clutch size (P = 0.237). No differences in C. aeneicollis egg clutch size or survival, nor P. melanderi egg presence. No differences in adult beetle abundance, or C. aeneicollis larval abundance either. In South Lake, as P. melanderi egg presence increases, C. aeneicollis egg clutch survival decreases.


Although these results would suggest that C. aeneicollis displays no biological differences at different elevations, whether be it egg clutch size or survivability, it may be that the sample size was just too small. This preliminary study does show that there is potential, and in the future, if all of the study sites originally intended for monitoring were actually productive enough to monitor, we may find more conclusive answers.
We were able to confirm that P. melanderi attacks significantly lowered the survival rates of C. aeneicollis as demonstrated before by other researchers (Rank and Smiley, 1994). This indicates that P. melanderi is a primary predator, yet even though none of these fly eggs were found at Big Pine Creek, the survival rate of the C. aeneicollis egg clutches did not differ significantly from the other drainages that had an abundance of this predator. One possible explanation may be that the especially warm spring season made the drainage less suitable for P. melanderi, but actually enhanced the habitat for other predators. According to Rank and Smiley, 1994 article, the flies were present previously, so for some reason, they just were not around this season. Our original theory was that the P. melanderi population would simply coincide closely with the C. aeneicollis population. Wherever we found the prey, the predator would also be there. However, this result clearly suggests that other factors play key roles also about which we do not know enough.
As for differences across drainages, aside from no P. melanderi at Big Pine Creek, everything was quite uniform. However, these results were not very conclusive and more work definitely needs to be done.


Table 1. Study sites at comparable elevations
Drainage Study Site Elevation (meters) Egg Clutches Marked
Rock Creek Heart Lake 3190 0
Mosquito Flat 3111 20
Pine Grove 2837 0
South Lake Pipeline 3131 30
Parking Lot 2996 30
La Hupp 2854 30
Big Pine Creek Upper Site 3229 30
Pond Bog 3128 30
Falls Site 2997 0

• (Map 1) 3 drainages

(Fig. 1) P. melanderi egg presence different in Big Pine.

o (Fig. 2) However, Big Pine Creek C. aeneicollis egg clutch survival is not higher than other drainage.

Within drainage
• (Fig. 3) In South Lake, as P. melanderi egg presence increases, C. aeneicollis egg clutch survival decreases.

About Us Science Fridays Staff Corie Cann: Bio Kimberly Ogden: References Kimberly Ogden: About Me