We will attempt to disentangle the complex allometric and biogeographic relationships between body size (volume), latitude, and various other traits in salamanders.

These relationships are illustrated below.

From this, we can come up with a series of simple linear models, excluding any interaction terms.

1) Vol~Gen
2) Vol~Area
3) Vol~Lat
4) Gen~MolRate
5) MolRate~DR
6) MolRate~Lat
7) DR~Lat
8) DR~Area
9) Area~Lat
Load the Data
setwd('~/Dropbox/Phylo_Comp_Meth/')
tree <- read.nexus('caud.ultra.nexus')
dat <- read.csv('Caudata_PhyloComp_Data.csv')
rownames(dat) <- dat$Species
dat <- dat[,c(1,8,10,13,14,15,16)]
dat <- na.omit(dat)

# Because these are raw data, we need to do a few things first. This includes converting latitude to absolute value (we are interested in the effects of distance from the equator), and normalizing non-normal data.
# Convert latitude to absolute latitude. 
dat$lat <- abs(dat$lat)

# Lets look at the normality of our data. 
hist(dat$cvalue.quant)

hist(dat$area)

hist(dat$DR)

hist(dat$Abs.Mol.Rate)