This may be a silly question to ask, but, is it possible do do a MANOVA-style analysis with a generalized linear model? I have a data set that I'm working with that, for each variable (time in this case, as it's a repeated measures MANOVA) is fit much better using glm rather than a traditional linear model - even after transforming the data so that the resulting (I'm log transforming the data for lm, whereas under glm the data is most appropriate for a model with a poisson error and an exponential link). However, under glm, if I try and fit my matrix of response variable, I get the error Error: (subscript) logical subscript too long Is there a library or method of analysis that handles this type of problem? Much obliged! -Jarrett ---------------------------------------- Jarrett Byrnes Population Biology Graduate Group, UC Davis Bodega Marine Lab 707-875-1969 http://www-eve.ucdavis.edu/stachowicz/byrnes.shtml [[alternative HTML version deleted]]