Background and aims – Many diatoms have spines on the surface of their valves. These structures differ from one taxon to another. Are all these spines the same? Are they homologues of one another? This paper sets out to explore some of the issues surrounding the determination of homologues with reference to members of Fragilariaceae.

Methods – A variety of spines from species in Fragilariaceae are examined (in the SEM) and position on the valve documented relative to those already recorded in the literature.

Key results – Spines that occur on the valves of some ‘araphid’ diatoms in Fragilariaceae can be interpreted in the light of where they are found. Spines that occur on the virgae can be thought of as modifications of that structure; spines that occur on the vimines can be thought of as modifications of that structure – the two kinds of spines are not homologues of each other. The term ‘spine’, on its own, is not useful for understanding taxon relationships; the term ‘spine’ is not even a character in the comparative biology sense but a descriptive catch-all for something that simply ‘sticks out from a surface’.

Conclusions – Systematic characters, those applicable to comparative biology, are modifications of other characters and so are, in one sense, like taxonomies: hierarchical. A consequence of this is that plotting morphological characters on molecular trees of relationships is a futile endeavour – treating characters and their modifications, as if they are static (unit) features of a non-changing entity, is book-keeping not science.