CHAPTER IV
SENSE ORGANS AND SENSES OF EARTHWORMS
As this is not an anatomical treatise we shall not attempt any detailed anatomical and histological account of the sense organs in this group of worms. But a few facts must be given in illustration and explanation of the senses of touch and sight that the Oligochaeta undoubtedly possess. These Annelids, unlike their allies the marine Polychaeta, and even their allies on two other sides, the leeches and flatworms, have no complexly organised eyes or other sense organs. They have in fact no organs to which a definite sense can be attached on histological grounds. There is nothing comparable to an eye or to the auditory sacs of other low worms. There are only particular cells of the epidermis, often associated into small groups, and those again into larger associations of rows of such groups of cells. It is to be presumed that these modified groups of cells have a sense function; but no more can be said than that they are doubtless tactile and also to some extent receptive of the influence of light. True visual cells have been asserted to exist in earthworms, consisting of cells of which a part is clear and transparent and has been supposed to serve as a lens for the rest of the cell which represents a retina. But belief in the function of these cells is by no means unanimous. On the other hand many investigations have proved the existence of groups of epidermic cells of an oval form which are at present arranged in definite rows upon the segments of both terrestrial and aquatic forms, which are moreover connected with nerve terminations, or are at least – according to more modern views – in close contact with the terminations of nerve fibres. These are furnished often at their free tips with minute sensitive processes. There is nothing in the structure of these to associate them definitely with any sense in particular. They suggest of course tactile organs more than organs of any other sense. In addition to these are certain problematical organs which are found in the Eudrilidae and are present in the members of one section of that group, the section which is represented by the universally found Eudrilus.
These bodies have been compared to a Pacinian body (a sense-organ found in Vertebrates) and they bear no little resemblance to it. For each consists of a darkly staining core surrounded by a layer of cells arranged like the coats of an onion. In any case it would appear that these bodies must be looked upon as of a sensory nature, though they do not reach the surface of the body but underlie the epidermis. Their function must remain purely a matter of guesswork at present, for nothing to the point is known of the habits of the Eudrilidae. It has been suggested by Dr Gustav Eisen that these cells are auditory and serve to warn the worm of the footsteps of birds and other enemies. That too is his view of certain peculiar but different cells found in the epidermis of Pontoscolex. In the latter something like an otosome has been found which is certainly lacking in the Eudrilidae, whatever may be the function of the cellular epidermic bodies here briefly referred to.
While there is thus nothing very positive to be gleaned from an examination of the structure of the Oligochaeta as to the senses which they may possess, experiment has done something towards an elucidation of their behaviour under stimuli and their reaction to light and to other forces which come into play during their lives. There is some evidence that earthworms can see, using that expression of course in a very broad way. At any rate they react to changes in the intensity of light. The gross experiment of flashing a lantern upon earthworms which are reaching out from their burrows with the tail end inserted in those burrows shows that they have an appreciation of light. More refined experiments have been conducted upon the sense of light. Dr Graber used a box with two compartments, the one of which was dark and the other illuminated with diffused daylight, i. e. not direct sunlight. The worms were allowed equally free access to both and were examined at the close of every hour, and their positions noted. The investigator found that on the average the dark half of the box contained 5·2 times as many worms as the light chamber, thus indicating a very marked preference for absence of light.
Not only is this the case, but the same observer proved that earthworms can distinguish between degrees of intensity of light. This obviously indicates a more complete 'visual' sense. He illuminated the light-box of the former experiment with light admitted through a ground glass screen, thus diminishing its intensity. The other chamber was left as before but the screen was removed, thus admitting full daylight. In this experiment the number of worms in each compartment proved to be different. The results were not so striking as before, since only rather more than one-half were found in the more dimly illuminated chamber. It is a well-known fact that if earthworms are abroad at all from their burrows, it is during the night that this movement takes place, the numbers decreasing towards morning though worms are often seen crawling about well after sunrise. Some experiments have been made in attempt of explanation of this apparent anomaly. It would appear from these experiments that while worms are negatively phototropic to strong and moderate light as has already been pointed out, they are positively phototropic to very dim light; hence the advent of evening calls them forth from their burrows. It will be noted that this perception is of very great advantage to the worm since its more active enemies above ground are diurnal. It was held originally that the head end of the worm only was thus sensitive to light; but more recent experiments have shown that this is not the case, and that all of the body is sensitive. This disposes of course of the existence of special light-receiving organs in the anterior part of the worm's body. Not only this, but an interesting extension of the view has been promulgated. It has been shown by Prof. G. Parker and a colleague that in the common Brandling worm, Allolobophora foetida, the response to light stimuli was related to the amount of the body exposed to its influence. This is very important as showing that the light perception in these creatures is probably not due to special organs having a limited position on the body, but is due to collective sense impressions of many cells scattered over the whole body, the impression being the greater when the whole body is exposed and less when only parts of it are exposed. Furthermore, and this has quite another importance, these observers noted that the reaction effects differed when only a part of the body was exposed; that they were greater in the front of the body, less in the middle, and less still at the tail end. Indeed they found that the reactions in the case of the front end of the body alone being exposed were rather more than one-third as compared with those which were shown when the whole body was subjected to the light stimulus. The fact that the least sensitive region of the body is the posterior end has, it is true, only been definitely proved in the case of the worm whose specific name has been mentioned. But it is possible that others are similarly affected. And it is highly important to note the prevalent habit among the Tubificidae of lying with the head end imbedded in the mud of the pool which they inhabit, while the tail end emerges and waves freely in the flood. The additional fact that this tail end occasionally bears gills (as in Branchiura sowerbii and Phreodrilus branchiatus) has a collateral importance not to be mistaken.
CHAPTER V
RELATIVE FREQUENCY OF EARTHWORMS INDIFFERENT REGIONS OF THE WORLD
It will be of use for various purposes to be considered later to arrive at a comprehensive view of the relative numbers of species and genera of earthworms in the four quarters of the globe. And in making this general census we shall not take into consideration the purely aquatic forms, but shall limit ourselves to the earthworms, sensu stricto, or Megadrili, of which, however, it is true that some members are actually lake and river dwellers. This latter fact will not, however, interfere with the usefulness of the comparative survey.
Two preliminary remarks are necessary. The opinions of naturalists vary as to the limits of genera; and a species may be a species to one and a mere variety to another. Thus it will be impossible to give a summary of the facts to be enumerated presently, which will be either absolutely accurate or which will satisfy everyone in every detail. But it is asserted that the following survey is substantially correct.
In the second place it is often possible to eliminate from the fauna of a given region those species and even genera which have been accidentally imported, a matter which will receive careful consideration on a later page. Such forms are therefore, in those cases at any rate where the evidence seems to be overwhelming, withdrawn from the list. In other cases, particularly in the Eastern region of the world, it has been found less easy to rectify the catalogues by removing what Dr Michaelsen has termed 'peregrine' forms.
We shall commence with a census of South America; the entire Continent will be divided for the present purpose into three divisions, viz. South America, Central and North America, and in the third place the West Indian Islands.
In South America we find that the bulk of the indigenous earthworms belong to the family Geoscolecidae and to a definite sub-family, viz. Geoscolecinae. These genera are Onychochaeta with one species, Hesperoscolex of which one species is known from the area, Periscolex with one species, Anteoides with two species, Pontoscolex one species. Opisthodrilus with two species, Rhinodrilus (including either as synonyms or as sub-genera, Thamnodrilus, Anteus, Tykonus, Urobenus and Aptodrilus) with no less than 49 species: Andiodrilus with five species, Holoscolex with one species, Glossoscolex ten species, Fimoscolex, Andiorrhinus and Enantiodrilus with one species apiece.
Thus of this sub-family of Geoscolecidae we have in South America a large number of genera and a much larger number of species. Of a second sub-family of Geoscolecidae there are three species of Criodrilus found in the South American Continent.
We now turn to the Megascolecidae of which a large number of species occur within the area now under consideration. The bulk of these belong to the sub-family Acanthodrilinae and they are as follows:
Of the genus Notiodrilus there are ten species, of Microscolex two species, of Chilota 19, of Yagansia 13.
A second sub-family Trigastrinae also occurs in this Continent and the following genera are permanent inhabitants, viz.: —
Dichogaster (= Benhamia) with only three species, of which two at least are found elsewhere, and of which therefore the autochthonism is doubtful.
Finally, there is the sub-family Ocnerodrilinae comprising the following genera: Kerria with ten species, Ocnerodrilus (with sub-genera Liodrilus, Ilyogenia and Haplodrilus) four species, which again are rather doubtful indigenes of the South American Continent.
Leaving aside certain species (of the genera Lumbricus, Pheretima, etc.) which are clearly not indigenous, the South American Continent harbours 150 kinds of earthworms which are distributed in some 19 genera. But of these a few species (e. g. Onychochaeta windlei, Kerria macdonaldi) stray into neighbouring regions, i. e. the West Indies and California. It is doubtful therefore whether they are to be referred to as limited to one of these regions and accidentally imported into the others, or whether they are genuine inhabitants of both.
The South American Continent shares with the West Indies the following genera, but the species (except in the case of Onychochaeta windlei, Glossoscolex peregrinus) are distinct; these genera are Hesperoscolex, Pontoscolex, Dichogaster, and Ocnerodrilus. Diachaeta is limited to the West Indies.
The following South American genera are also found in Central and warmer North America (Mexico, California), viz. Hesperoscolex, Periscolex, Rhinodrilus, Pontoscolex, Andiodrilus, Glossoscolex, Notiodrilus, Microscolex, Dichogaster, Kerria and Ocnerodrilus. But with the exception of Hesperoscolex, which seems to belong rather to Central America and the West Indies, Microscolex and Pontoscolex which are world-wide and whose original home is therefore difficult to fix, and Dichogaster and Ocnerodrilus which would seem rather to be rare immigrants (perhaps not truly indigenous) into South America, these genera are practically distinctively South American.
Thus we may fairly say that the genera Anteoides, Opisthodrilus, Andiodrilus, Holoscolex, Glossoscolex, Rhinodrilus, Andiorrhinus, Fimoscolex, Enantiodrilus, Notiodrilus, Chilota, Yagansia and Kerria are at least very distinctive of S. America and that they are represented by the large majority of species found in that continent, the total being 120 or slightly more.
Leaving the American Continent and adjacent islands and archipelagos, the next great land-mass to receive attention will be the Continent of Africa. In giving a census of the species and genera of earthworms which inhabit this quarter of the globe it must be premised that the facts relate only to Africa south of the Sahara. But little is known of the genera which occur in Algeria and Morocco, but from what little is known it is clear that they should come into consideration in connection with the fauna of Europe and not with that of tropical and south temperate Africa.
We have in the first place to consider an entirely peculiar family, of fair extent in genera and species, which is limited to this region of the world; that is to say with one apparent exception which is clearly only apparent. The genus Eudrilus is one of the few kinds of worms that turns up in collections from every tropical part of the world and even at times from more temperate countries. It is one of those 'peregrine' forms, as Dr Michaelsen has termed them, which possess unusual facilities for extending their range. Presumably its real home is Africa. This family is known as the Eudrilidae though by some it is only regarded as a sub-family of the Megascolecidae. In this family we have the following genera: Eudriloides with 11 species, Platydrilus with 11 species, Megachaetina with two species, Reithrodrilus with one, Stuhlmannia with five species, Notykus and Bogertia with one species each, Metadrilus also with but one species, Pareudrilus with perhaps five, Libyodrilus with one and Nemertodrilus with two species, Metschaina with two species, Eudrilus with two (or possibly more?) species, Parascolex with four species, Preussiella with two, Buttneriodrilus with two, Eminoscolex with 16 species, Hyperiodrilus, Heliodrilus, Alvania, Rosadrilus, Kaffania, Euscolex, Metascolex, Beddardiella, Gardullaria with only one species to each genus, Bettonia with three species, Teleudrilus with 15 species, Polytoreutus with 22 species, Iridodrilus with two, Malodrilus also with two, Neumanniella with eight, Eupolytoreutus with two and Teleutoreutus with only one species.
This completes the list of the Eudrilidae. We will take the huge family Megascolecidae next; and we find in tropical Africa that the genus Dichogaster alone contains at least 93 species confined to Africa, as well as a few which it shares in common with America, and the common and widely spread D. bolavi which has even made its way to Europe. Of the sub-family Ocnerodrilinae we have the genus Gordiodrilus with seven species, a genus which also occurs in Madagascar and the West Indies: Nannodrilus with three species, Diaphorodrilus with one species and a few examples of Ocnerodrilus and its sub-genera, some of which are also forms that occur elsewhere in the world, for example Nematogenia panamensis. Of Pygmæodrilus there are eight or nine species.
Of the sub-family Acanthodrilinae the Cape region of South Africa harbours some seven species of the genus Notiodrilus, of which one however is a West African form. The allied Chilota has 13 species, and there is a peculiar genus Holoscolex near to Yagansia with one species. We next have to deal with the Geoscolecidae, of which a sub-family, Microchaetinae, is mainly found in Africa, the rest being found in the neighbouring Madagascar and some few in the East. Microchaetus contains about 14 species, Tritogenia perhaps three, Callidrilus two, and the genus Glyphidrilus, mainly found in Asia, has one species in the region now under consideration. In addition to these Geoscolecids there is the peculiar and largely aquatic Alma with six or seven species in East, West, and Central, Africa, and in the Nile region.
Summing up the genera which are found in tropical and South Africa we find that there are 44 which are abundant in, or entirely confined to, the region. In addition to these four or so occur in Africa but are either more abundant elsewhere or (as in the case of Chilota and Notiodrilus) are equally distinctive of other parts of the world. The number of species may be estimated at 270, possibly rather more. Clearly therefore this part of the world is much richer than South America, both in numbers of genera actually found, and peculiar to the country, and numbers of species.
Passing from Africa the next definite quarter of the globe which will detain us here is the island of Madagascar, so remarkable for the Mammalian fauna which characterises it, for its lemurs, peculiar Insectivora and Carnivora, and above all by reason of the absence of the prevalent African types such as antelopes, zebras, rhinoceros etc. It is probable that a good deal still remains for discovery among the earthworms of this island; but a considerable number are already known and they are as follows:
The Eudrilidae are completely absent, a state of affairs which is paralleled by the absence of antelopes among mammals.
The sub-family Acanthodrilinae of the family Megascolecidae are represented by four species of the genus Notiodrilus, with which perhaps Maheina braueri from the Seychelles should be included as it presents small differences from Notiodrilus.
Among the Megascolecinae a good many species of Pheretima have been collected both on Madagascar and on certain adjacent islands; but these, with one possible exception, are forms which occur elsewhere and are often indeed very widely distributed 'peregrines,' so that it is hardly permissible to place them among the real inhabitants of Madagascar. The same arguments hold in the case of Lampito mauritii and the ubiquitous Eudrilus and Pontoscolex. But there is the peculiar Howascolex.
Among the Ocnerodrilinae a distinct species of Gordiodrilus occurs, and an obviously introduced Ocnerodrilus.
It is among the Geoscolecidae that the most characteristic forms are met with. These belong exclusively to a genus of the Microchaetinae, Kynotus, which is found nowhere else but in Madagascar, where it is represented by at least twelve different species.
We have therefore in Madagascar and the surrounding islands only two peculiar genera, only four genera which can be regarded as endemic, and only about seventeen peculiar species.
Passing eastward and crossing the Indian Ocean we come to the Continent of Asia, and we shall first direct attention to the peninsula of India and adjacent parts of Burmah and the island of Ceylon, of which there has been accumulated a great deal of knowledge concerning the Oligochaetous fauna.
This quarter of the globe differs quite as much from any that have been hitherto considered as they do among themselves. We have left behind us the Geoscolecidae with the exception of the ubiquitous and clearly peregrine Pontoscolex, and the genus Glyphidrilus, which, being aquatic at times, is perhaps hardly to be considered in the present survey.
We have also in this Indian region the equally ubiquitous Eudrilus eugeniae which need not of course detain us further. The Lumbricidae are for the most part of European forms with the exception of Helodrilus indicus conceivably an actual inhabitant of India. It is among the Megascolecidae that the vast majority of the forms endemic in India are to be found. We shall take this family according to its sub-families. In the first place we note that the sub-family Acanthodrilinae is totally unrepresented. The large sub-family Megascolecinae has very numerous representatives. Of the genus Megascolex itself there are some 30 species, of which, however, two or three are looked upon as varieties. Of the allied genus Notoscolex there are ten species and of Perionyx about 13. The genera Megascolides, Diporochaeta, Spenceriella and Woodwardia have only six species between them of which three belong to Megascolides. Plutellus has five species in this region. Lampito, which is a widely spread form with but one species L. mauritii, may or may not find here its original home. It also occurs in Africa and Madagascar. Pheretima is represented by no less than 12 species, of which at any rate the very great majority are peregrine forms, and not to be safely regarded as forming an integral part of the fauna of the Indian peninsula and adjoining countries.
Of the sub-family Octochaetinae the type genus Octochaetus has ten species in this region and Eutyphoeus, which is restricted to the region, has 15 species or perhaps rather more. Hoplochaetella has but one. Of the Trigastrinae there are four or five examples of the genus Dichogaster which occur within the region now under consideration; but it is doubtful whether they are truly indigenous. On the other hand Eudichogaster, closely allied both to Dichogaster and Trigaster, seems to be confined to this part of the world where it is represented by five species. Finally we come to the last sub-family – that of the Ocnerodrilinae, which is represented by a species apiece of the genera Ocnerodrilus, Nematogenia and Gordiodrilus. The latter species G. travancorensis is alone to be regarded as endemic and it is very near to the African G. zanzibaricus, described some years since by the present author.
The remaining family of terricolous Oligochaeta found in India is the family Moniligastridae which is practically limited to this part of the world and consists of at least twenty species distributed among the genera Moniligaster, Eupolygaster, Desmogaster and Drawida, the majority belonging to the last-named genus.
This quarter of the globe is therefore inhabited by 18 genera which are certainly truly endemic, and which comprise between them about 120 species. But only four or five genera are peculiar.
The remainder of the Asiatic Continent is not very well explored with regard to its earthworm inhabitants. It seems clear however that the southern and coastal region of China and Japan with the Malay peninsula are really continuous with the mass of islands which lie between India and Australia and form together a tract of land which is characterised by an Oligochaetous fauna differing from both that of India on the one hand and Australia on the other. We shall therefore consider this huge portion of the globe as one region comparable to the other divisions that have been hitherto considered. With reference to the Lumbricidae and Geoscolecidae the same remarks may be made as in the case of India. The indigenous forms of the latter family are to be looked upon as outside of the present survey since they are largely or entirely aquatic forms. Pontoscolex corethrurus, and Eudrilus eugeniae, need not detain us for reasons already amply stated. We now come to the great family Megascolecidae. Of this family the genus Pheretima stands at the head; and of the 200 or so species that have been or can be assigned to this genus all, with merely two or three exceptions, are natives of the Eastern Archipelago and adjoining mainlands of Asia. Of other Megascolecinae the region has yielded the following genera. In Java one species of Woodwardia (W. javanica) has lately been described. It is regarded by Michaelsen however as doubtfully indigenous. And the same remark may be made of Perionyx. Plionogaster, however, with four or five species, is limited, as far as our present knowledge goes, to the Philippines and to neighbouring islands. The Acanthodrilinae and Octochaetinae are totally absent from this part of the world, there being no record even of peregrine species of these sub-families. The sub-family Trigastrinae is not however unrepresented; for of Dichogaster several species occur, such as D. malayana and D. saliens. But these are by no means certainly to be looked upon as real natives of the situations within this area where they occur. This completes the scanty list of genera found in the region under consideration; for one Moniligastrid (M. barwelli) is hardly to be looked upon as indigenous. We have therefore to record here but five genera, of which only two are certainly indigenous and probably also confined to the region; they contain between them certainly two hundred species.
We next come to the Continent of Australia. The earthworm fauna is again quite without Geoscolecidae and of course Lumbricidae. As to the former there are not even doubtful cases like Glyphidrilus of the East; for we find only recorded Pontoscolex corethrurus, a species concerning whose extraordinary powers of migration there is no possible doubt. The same may be said of Eudrilus also recorded from Australia. Here the Megascolecidae are as conspicuous as in the old world generally. We find, however, a great many members of the sub-family Acanthodrilinae. Of the antarctic genus Notiodrilus there are some six species. Microscolex occurs; but the real habitat of this genus is very doubtful. Diplotrema, with one species, D. fragilis, is not only indigenous to, but confined to, Australia.
The most prevalent sub-family is that of the Megascolecinae. Of the genus Pheretima there are two species which may or may not be truly indigenous. Of Plutellus there are forty species; there are over thirty of Diporochaeta, while Notoscolex (with which Dr Michaelsen associates Digaster, Didymogaster, Perissogaster) includes more than forty species, Fletcherodrilus has but one species, Megascolex has seventy species, Woodwardia sixteen, Spenceriella five, and Megascolides seventeen. There are four species of Perionyx which were at one time regarded by Michaelsen as necessitating a new but allied genus Perionychella; the two are now merged. Thus there are not far short of 150 species and eleven genera represented, of which only two are limited to Australia.
Having completed the survey of the central and southern land masses of the globe we next direct attention to the northern land masses, viz. North and Central America on the one hand, and Europe and northern Asia on the other. With regard to America we find besides many species of Lumbricidae the genera Notiodrilus, Microscolex among the Acanthodrilinae, Megascolides and Plutellus of the Megascolecinae, no member at all of the Octochaetinae, a considerable number of species of Ocnerodrilus and sub-genera belonging to the Ocnerodrilinae, a good many species of Dichogaster and at least one of Trigaster among the Trigastrinae, while one sub-family, that of the Diplocardiinae, is only found here and contains two genera Diplocardia and Zapotecia with quite ten species between them of which only one belongs to the last-named genus. There are no Geoscolecids (except the chiefly aquatic Sparganophilus); this family stops short in the West Indies where their presence has been briefly referred to in considering the worms of South America. The vast majority of the genera enumerated here are only found in the warmer parts of the North American continent. We have therefore in this division of the world some nine genera of which at least one, viz. Diplocardia (and its close ally, hardly perhaps to be separated, Zapotecia), is confined to it, while Trigaster only extends as far south as to the West Indies.
Europe and northern Asia, of which Europe alone and certain limited tracts of Asia are at all known, contain all the genera of the family Lumbricidae which are (according to Michaelsen) Eiseniella, Eisenia, Helodrilus, Octolasium, Lumbricus with various sub-genera of Helodrilus, such as Dendrobaena and Allolobophora. Besides these forms, which amount to at least 130 species, we find the genus Hormogaster, with two species, the only genus of the Geoscolecid sub-family Hormogastrinae; there is also a species apiece of the genera Criodrilus and Sparganophilus (sub-family Criodrilinae). But as these are at least largely aquatic they come under another set of rules and are not important in the present survey of the earthworms of the world. It will be thus seen that with these two last exceptions the area in question contains but six genera of which all are peculiar.
We have now completed the survey of the principal land-masses of the globe. We shall conclude with a reference to one of the largest islands of the world, viz. New Zealand, with which will be included a few outlying islands such as the Aucklands, Snares Island, etc. The reason for not ignoring these islands as we have ignored certain other islands of large size, such as Japan, will be apparent from the peculiarity of the earthworm fauna which they show. From New Zealand the Moniligastridae, Geoscolecidae, Lumbricidae, and, of course, the Eudrilidae, are absent, save the ubiquitous Eudrilus. The only family of earthworms which is here represented is that of the Megascolecidae. Of this family the Acanthodrilinae are very well represented. We have at any rate seven species of the genus Notiodrilus of which the bulk are from the small adjacent islands and not from the mainland. The genus Rhododrilus with nine species is actually limited to the New Zealand group and so is an allied genus consisting of two species only, viz. Leptodrilus. This latter genus is confined, so far as present information goes, to the Auckland and Campbell islands. Dinodriloides with two species is also limited to New Zealand and to the North Island. Maoridrilus with some ten species is another native and restricted genus. So too is the allied Plagiochaeta with numerous setae on each segment but with the alternating and single nephridial pores of Maoridrilus. There are several species of Plagiochaeta of which one has been lately regarded by Dr Michaelsen as really falling within the otherwise Indian genus Hoplochaetella, while for another he has formed the genus Pereiodrilus. Neodrilus with but one species N. monocystis is another peculiar New Zealand genus. The family Octochaetinae contains only four genera, of which one, viz. Dinodrilus (with four species), is limited to New Zealand, while Octochaetus has about five representatives. Hoplochaetella (if Dr Michaelsen's surmise referred to above be correct) has one species in New Zealand. The Megascolecinae are less numerous than the Acanthodrilinae, but there are ten species of the genus Megascolides (which includes Benham's genus Tokea), perhaps seven species of Diporochaeta, and two other species which Michaelsen has removed from the genus Diporochaeta and placed in Spenceriella. One Plutellus (which however may have been introduced) completes the New Zealand Megascolecines. We have therefore in this part of the world fifteen genera including between them some 58 species; eight of the genera are peculiar to the islands.