ARTICLE : Digital ID

 Digital ID

The characteristic of soral position – which choice matches your fern?

While trying to identify a plant using a traditional binary key (as in the Flora of New Zealand series), have you ever been frustrated by coming across a couplet that you cannot answer with your specimen?   For example, “Fruit red - go to 2; fruit blue - go to 3”, and your specimen, while with wonderful flowers, has no fruit!   I’ve often been discouraged by such stumbling blocks, and I also don’t enjoy having to look in a glossary to understand every second word used in a key.   However, plant identifications may be about to get easier, at least for the ‘computer literate’.

Various computer programmes (e.g., DELTA, LUCID) now allow for the construction of so-called “interactive keys”.   These operate by presenting a suite of plant characteristics (e.g., leaf hairiness), each of which has at least two states (e.g., hairy all over; hairy on the margins only; not hairy).   By selecting on the computer the state shown by your specimen, you eliminate all of the species without that state from consideration (e.g., by selecting “not hairy”, the computer now knows that your specimen cannot be a hairy-leaved species).   By repeating this for other characteristics, you can progressively whittle away all of the potential candidates.

A great advantage of interactive keys is that you are not constrained in the order you process your specimen’s characteristics.   For instance, all is not lost if your plant is without flowers, as you can work through the leaf characteristics.   Furthermore, in this digital format, illustrations can be plentiful.   If, for example, you are not sure what the character “soral position” refers to, then the supporting pictures should make everything clear; simply choose the picture option that matches your specimen.   If you are still puzzled, ‘notes’ (with technical terms explained) that help you make a choice are just a mouse-click away.   And if that also fails, you can still retreat to try a different characteristic that you are more comfortable with.   The usual interactive key holds a large number of characteristics with which to compare your specimen, but you usually only need to answer a small subset of them to get an identification.

Often you can progress until only a few or even just one species remains.   At this point you can view pictures and other information (e.g., notes about how to distinguish similar species).   This can help further narrow the field of candidates if necessary, or if only one species remains you can check whether your identification is correct.   Of course, you can wade into this information on each species right at the beginning, in the same way you might flick through a book’s pictures, but that would not be using these keys to their full advantage.   The wonders of interactive keys are, however, not an excuse to collect poor specimens; good samples with mature reproductive structures always make identification easier.

If you’ve used the interactive key to eliminate all candidates but Hymenophyllum flabellatum, you can check the pictures to confirm that your specimen really is that species.

Two DELTA interactive keys are under development at Te Papa.   Patrick Brownsey is building one for New Zealand’s ferns, and Mike Bayly and Alison Kellow for Hebe.   The creation of these keys is not a trivial task.   Firstly, one has to have a good idea of the species that are ‘out there’, and how they can be distinguished from one another.   Then this data must be coded into the software.   Obtaining useful pictures is also a big job, and both the fern and Hebe keys have benefited from Bill Malcolm’s considerable imaging skills shown here.

Both keys have been trialled by willing guinea pigs from Wellington Botanical Society in the last year or so, with about 15 members attending each of the Hebe and fern workshops.   On both occasions, Phil Garnock-Jones generously arranged access to one of Victoria University’s computer-equipped biological laboratories.   Botanical society members appeared to enjoy the new technology, while the developers of the keys benefited from the identification of errors in the data and matters of user-friendliness.

Exactly how, or when, these keys will be available is yet to be decided; possibly via CD, or even accessible through the web.   Computer-based tools are admittedly not ideal for use in the field, but then I don’t know anyone who lugs around the Flora series on tramps either.   However, I do look forward to a future where a device the size of a mobile phone holds all the botanical information you could want, including these image-rich interactive keys, as well as topographic maps and a GPS function.   Wouldn’t that be great for botanical explorations?   The technology is probably not too far away, but, speaking from near the coalface, producing illustrated interactive keys for the entire flora is going to take a lot of work!

Leon Perrie