A Few Words About KEYS

Adapted from Harold Oldroyd 1958.

Keys work by a process of elimination, gradually narrowing down the number of possibilities. It is important to understand that a key is much more trustworthy in proving that your insect is not A, than in proving that it is B. It might, for instance, actually be C, a species not mentioned in the key. A key does not prove anything positive, it only suggests possibilities. People used to be faddy about the sort of key they liked, but these days the arrangement that has the greatest acceptability for simplicity and directness is the dichotomous key, so-called because at each step it asks you to choose between two alternatives. Thus:

1..... Two pairs of membranous wings...........................................2

Only one pair of membranous wings, the other pair

being either hardened into wing-cases, or absent ..................29

2...... Fore-wings and hind-wings alike............................................3

Fore-wings and hind-wings different......................................I8

3..............................................(and so on).

If the insect agrees with the first alternative of couplet 1, then you procede to couplet 2; if the first alternative of couplet one is wrong then read the second alternative, if this is correct, then you procede past all the intervening couplets, and read on at couplet 29 as indicated. Go on like this until you come to a final choice where you will be offered a probable identification:

i.e.

35.... Abdomen with forceps......................... Earwigs (Dermaptera)

Abdomen without forceps...................... Beetles (Coleoptera)

If you are lucky this will give a correct identification, but it may not. Using keys appears difficult at first but it becomes easier the more you practice and is quite rewarding. A good way to understand keys better is to make your own. Remember that the details mentioned in the key are not the real reasons why we believe an insect is any particular species: they are only pointers towards something we can recognise. The real identification is made by comparing the specimen with another specimen that is already named, or with a drawing, or with a good, detailed description. however many modern keys contain, or are appended by a sufficiently good description to allow for competant identification.

People commonly make one of two mistakes in using keys. The first is to be over-cautious, taking every word literally, and refusing to go on if the specimen does not agree exactly. Remember that insects, like humans, have their individual variations. Think of making a key that tried to classify the people present at one moment at Waterloo Station, and group them into rigid categories, without any margin of error. If you come to a point in a key at which you cannot decide which of the two alternatives fits the specimen, do not give up. Consider first if it definitely fits neither: for example, if both alternatives talk about wing-structure, and your specimen has no wings, then the likelihood is that you have reached a wrong part of the key; or else your specimen is an exceptional one that the key does not properly provide for. If, on the other hand, either alternative could be considered to fit, try each one in turn and see how you get on. If the key is a good one you will usually find that on one of the two paths you quickly gain confidence again. Here again, experience of the group will help you to know which characters mentioned are likely to be clear-cut, and which you would expect to find very variable.

Making Your Own Key

Generally speaking, the most diffficult part of a key to construct is the beginning, separating the bigger groups. Obviously, the bigger the group, the greater the variation that is likely to be found in it, and the more exceptions there will be to any definite statement that one may make about the group. Consequently, when an insect is being identified by means of a key, the first two or three steps are likely to be the most uncertain. Fortunately, experience redresses the balance, because after a little practice you will know at sight more or less where the specimen belongs, and will need to use only the last few couplets, where the differences are more definite. The second fault in using keys is to trust them too implicitly, to run down the specimen rather casually, and then to label it and put it away in a collection without checking it further. Many errors have arisen through this practice, especially when the misidentification has been published. Do remember that the key does not identify the specimen: it only gives a hint of what it might be. The correct attitude to keys--at least to keys made by other people--is to be sceptical without being defeatist. The best keys are often our own because when we read a couplet in a key we ourselves wrote, we have the advantage of a great many mental impressions that are not mentioned in the couplet, and which are not available, of course, to anyone else. That is why so often the only person who can use a particular key successfully is the author.

The arrangement of couplets illustrated above is the most common, and to my mind incomparably the best one. The advantage of having the two alternatives side by side, so that you can weigh up your specimen against each of them, far outweighs the benefits claimed for other arrangements. Many entomologists, particularly those of an older generation, prefer the following style:

01. (I6) Two pairs of membranous wings ...

02. ( 9) Fore- and hind-wings alike ... ...

03. ( 7) . . .

09. ( 2) Fore-wings different from hind-wings

I6. ( I) Only one pair of membranous wings...

           . . . (and so on)

If 1. is correct you proced to 2. but if it is incorrect you procede to16 to read what would have been the second alternative in a dichotomous key.

This kind of key is often significantly called a 'Table', and is said to show the grouping of the insects better, while the numbers in brackets, referring to the other half of each couplet, make it easier to trace a way through the key, both forwards and backwards. Bracketed numbers can, of course, be used in the other type of key to help in retracing one's way. The need to turn over, sometimes, many pages to find the alternative character in a 'Table' of this kind, makes the operation into too much of a hurdle race.

Beware of keys in which more than two alternatives are presented at one time, especially if the author does not bother to indicate this very clearly. This is bad practice, because it defeats the first object of the dichotomous key, which is to proceed by a series of clear choices between simple and mutually exclusive alternatives. Occasionally, it is true, the insects fall naturally into three or more co-equal groups: e.g. if the legs can be red, or black, or yellow. But such a choice can nearly always be set out in couplets with a little more trouble, and the slightly greater printing space is more than compensated for by the assistance it gives the reader.

Another thing to look out for in keys--and to avoid in your own--is 'leapfrogging', like this:

40. Femora greatly swollen......................................41

Femora slender..............................................72

4I. Femora with spines beneath...............................71

No ventral spines on femora...........................85

42. Eyes hairy ........................................................48

Eyes bare.......................................................43

Of course, if you read this carefully, you will have no trouble, but keys are meant to give help in quick identification, and you should always try to keep to the logical order. Remember keys should be designed to be used by someone with no previous experience of the species or groups involved. The construction of keys is a very searching test, not only of how much you really know about a group of insects, but also of whether your knowledge has been assimilated by a tidy and logical mind. When an author publishes a complicated and disorderly key, you will find that the rest of his work shows evidence of a chaotic and turbulent state of mind.

Constructing and Adapting Keys

For all that we have just said, do not think that making keys is only for the professional entomologist. If you have studied a group of insects, it is well worth while to make a key to them, for your own use, even if you have no idea of publishing it. It is even more useful to be able to modify an existing key and adapt it to your needs: for example, to take a key to the European genera and species of a family and to abstract from it those which occur in Britain; or to combine two or more short keys into one that is more comprehensive. You may need to do this if you do not own the books concerned, but have to borrow them from a library, and return them after a limited time. Then you can pick out just the parts you need, without having to copy out a great deal that is of no value to you. For all key-making, I suggest you use a system based upon a card-index, using standard cards 5 X 3 in. Suppose we begin by copying a short key from a volume borrowed from a library. Write each couplet on a separate card, and stack the cards in a filing cabinet, or in one of the small portable index-boxes (called, for some mysterious reason, a 'trial outfit'); or at worst, hold the cards together with a rubber band. Then if you want to add more names to the key, run each one down as far as it will go, and add another card at that point, using a decimal system of numbering, thus:

The use of cards avoids all the crossing out, cutting-up and pasting together that is needed if you copy the oiginal key on to sheets of paper. Such a card-key can be expanded indefinately, each timejust for the labour of adding a new card, an making a small alteration to the previous one (for this reason it is often better to write in pencil rather than in pen). To put a card between 21.5 and 22 you may number it 21.6; if you need one between 21.5 and 21.6 then it becomes 21.55, and so on like the Dewey Decimal System used in libraries. So long as the key is only for private use, and is liable to be altered, the decimal numbers can remain. When you want to publish it, you go through from the beginning, numbering the cards in order in red ink and, of course, not forgetting to make the corresponding alterations to the right-hand numbers.