Post date: Apr 03, 2017 3:54:29 PM
CHC could causally affect desiccation tolerance. I want to conduct a desiccation tolerance experiment to quantify variation in desiccation for the same populations (and perhaps families) that we will have CHC data for. I tried a small trial to get the methods figured out on March 31st 2017.
I used four caterpillars from each of two populations (BMT and SKI) (eight total). Caterpillars were weighed twice on the high precision scale (this took about 25 minutes; 8:50-9:14). I then put them on the benchtop under a pair of lights in petri dishes with no food (9:23).
I weighed them again at 10:17 (about 1.5 hours after the first weight was done). This is time point 1 (again with two reps). At this point, half were transferred to a container with desiccant. For these the petri dishes were opened to the air (gauze stuff on top). The other half were kept as is and both put under the lights (this was all set back up by 10:40).
Weights were taken again at 12:15 (done at 12:40) and caterpillars were put back under the lights (time 2). Two were dead.
A single final weight was taken at 3:15-3:25 (time 3). Six of the eight caterpillars were dead.
The data are here (trial data).
Weight loss was substantial (~10-80% of initial weight was lost). There wasn't an obvious effect of treatment (I can probably not worry about using the desiccant, they dehydrate just from the lights and lack of food/water). And, there was essentially no evidence of weight loss from frass (they didn't really produce any). Interestingly, even with n=4, there was a significant effect of populations (more weight loss by g or % for SKI than BMT). This is really encouraging.
Here is some R code I used to analyze the data from the on-line spreadsheet (below), and here are the main graphical results: (change over time), (total change).
dat<-read.csv("Downloads/TrialDesiccationExperiment - Sheet1.csv")
dat<-dat[1:8,1:10]
cs<-c(rep("blue",4),rep("orange",4))
ll<-as.numeric(dat$Treatment)-1
x<-c(0,0,1.5,1.5,3.5,3.5,6.5)
pdf("desTrialChngTime.pdf",width=5,height=5)
par(mar=c(5,5,0.5,0.5))
plot(x,as.numeric(dat[1,3:9]),ylim=c(0,0.033),type='b',col=cs[1],lty=ll[1],ylab="weight (g)",xlab="time (hours)",cex.lab=1.4)
for(i in 2:8){lines(x,as.numeric(dat[i,3:9]),type='b',col=cs[i],lty=ll[i])}
legend(0.2,0.01,c("BMT-des","BMT-con","SKI-des","SKI-con"),col=c("blue","blue","orange","orange"),lty=c(2,1,2,1),pch=21)
dev.off()
pdf("desTrialTotChng.pdf",width=5,height=8)
par(mfrow=c(2,1))
par(mar=c(5,5,0.5,0.5))
barplot(abs(dat[,9]-dat[,3]),col=cs,ylab="weight loss (g)",cex.lab=1.4)
legend(0.5,0.02,c("BMT","SKI"),fill=c("blue","orange"))
barplot(abs(dat[,9]-dat[,3])/dat[,3],col=cs,ylab="% weight loss",cex.lab=1.4)
dev.off()