At each time point t, you observe N_{i,t} spores from individual i, who is at a certain maturity level. These spores are of four types, and their counts are [n_{i,t,1}, n_{i,t,2}, n_{i,t,3}, n_{i,t,4}], where \sum_{k=1}^4 n_{i,t,k} = N_{i,t}. You expect that the prevalence of spore types varies with maturity. I think you may want to look into multinomial regression.