e in octave

[james mcskimming]

i've always accepted that e=exp(1) is a constant available in Octave
just like pi and others.  But i wrote code using e as a constant only to
discover it didn't work when a friend tried to run it on Matlab.
I've never used Matlab but assumed there would be compatability at such
a basic level.
Is there a reason why? and other than defining the variable e=exp(1) in
code for Matlab, is there a better way?
thx
Jas