wolffd@0: Cell arrays are a little tricky in Matlab.
wolffd@0: Consider this example.
wolffd@0: 
wolffd@0: C=num2cell(rand(2,3))
wolffd@0: C = 
wolffd@0:     [0.4565]    [0.8214]    [0.6154]
wolffd@0:     [0.0185]    [0.4447]    [0.7919]
wolffd@0: 
wolffd@0: C{1,2}  % this is the contents of this cell (could be a vector or a string)
wolffd@0: ans =
wolffd@0:     0.8214
wolffd@0: 
wolffd@0: C{1:2,2} % this is the contents of these cells - returns multiple
wolffd@0: answers!
wolffd@0: ans =
wolffd@0:     0.8214
wolffd@0: ans =
wolffd@0:     0.4447
wolffd@0: 
wolffd@0: A = C(1:2,2)  % this is a slice of the cell array
wolffd@0: ans = 
wolffd@0:     [0.8214]
wolffd@0:     [0.4447]
wolffd@0: 
wolffd@0: A{1}  % A is itself a cell array
wolffd@0: ans =
wolffd@0:     0.8214
wolffd@0: 
wolffd@0: 
wolffd@0: 
wolffd@0: >> C(1:2,2)=0 % can't assign a scalar to a cell array
wolffd@0: C(1:2,2)=0
wolffd@0: ??? Conversion to cell from double is not possible.
wolffd@0: 
wolffd@0: 
wolffd@0: >> C(1:2,2)={0;0} % can assign a cell array to a cell array
wolffd@0: C(1:2,2)={0;0}
wolffd@0: C = 
wolffd@0:     [0.4565]    [0]    [0.6154]
wolffd@0:     [0.0185]    [0]    [0.7919]
wolffd@0: 
wolffd@0: 
wolffd@0: BTW, I use cell arrays for evidence for 2 reasons:
wolffd@0: 
wolffd@0: 1. [] indicates missing values
wolffd@0: 2. it can easily represent vector-valued nodes
wolffd@0: 
wolffd@0: The following example  makes this clear
wolffd@0: 
wolffd@0: C{1,1} = []
wolffd@0: C{2,1} = rand(3,1)
wolffd@0: 
wolffd@0: C = 
wolffd@0:               []    [0]    [0.6154]
wolffd@0:     [3x1 double]    [0]    [0.7919]
wolffd@0: 
wolffd@0: 
wolffd@0: Hope this helps,
wolffd@0: Kevin