Each of the user-supplied list of datasets is checked against the first to see that its structure is similar. The output dataset is formed using the total lengths of the axes found in the input datasets.
The datasets are merged in the order in which they are entered.