March 30, 1871] 
NATURE 
437 

bution has {been eftected by land-continuity, we should expect 
that the proportion of winged and apterous species that extend 
their range beyond the island, should not be very strikingly dif- 
ferent from the proportion that is found on the island. We do 
not find, for example, that the proportion of the win less Caradi 
that have reached our own country from the continent by former 
land connection, is very different from that of the winged 
Cicindéle. 
Now, leaving out altogether those species which have certainly 
been introduced by man, and grouping the remainder for conve- 
nience in six divisions, we find that the Madeiran Coleoptera, 
which are not peculiar to it, may be classed as follows :— 
31 species of Carabide, of which 26 are winged, 5 apterous. 
The whole fauna, however, presents the very different 
proportion of 38 winged, 43 apterous. 
93 species of the families from the Hydradephaga to the 
Tomicidz inclusive, of which 90 are winged, 3 apterous. 
Total fauna ; 220 winged, 27 apterous. 
28 species of Curculionidae, of which 26 are winged, 2 ap- 
terous. Total fauna; 35 winged, 74 apterous. 
15 species of Longicornia and Phytophaga, of which 15 are 
winged, none apterous, Total fauna: 48 winged, 1 
apterous. 
20 species of Heteromera, of which 16 are winged, 4 apterous. 
Total fauna: 28 winged, 27 apterous. 
76 species of Staphylinidz, of which all are winged, none 
apterous. Total fauna: 109 winged, 6 apterous. 
The totals are, for the wide-ranging species, 249 winged, 14 
apterous = 263 ; for the whole fauna, 478 winged, 178 
apterous = 656. 
It thus appears that, in every case, an immensely smaller pro- 
portion of apterous than of winged species are widely dis- 
tributed. If we take the totals, we find that while about 
two-fifths of the whole number of species range to other countries, 
only about one-thirteenth of the apterous species do the same, 
although among the strictly endemic species there are 160 
apterous to only 110 winged! We can hardly impute such a 
constant and overwhelming preponderance to the fact that apterous 
insects have less facilities for extending their range, when we 
know that nearly every apterous genus possesses species of 
almost universal European distribution. I may here recall the 
fact, that of the above-mentioned fourteen apterous species which 
range out of Madeira, two are Meloes and two ants’-nest beetles, 
whose presence we have already sufficiently accounted for. It 
may no doubt be said that much of the difference here shown is 
due to the fact that the peculiar Madeiran species have had time 
to become apterous, while the species common to other countries 
have not yet had time to lose their wings; but this argument, 
although a valid explanation of some portion of the facts, if we 
admit that many of the latter have been recently introduced by 
natural causes, cannot be used by those who maintain a former 
land-connection as the sole origin of the fauna; for on that 
theory all the species now inhabiting the island (and not intro- 
duced by man) must date back to the same remote period, and 
have had equal time in which to be modified. 
Let us now consider what are the special relations of the 
apterous Madeiran species as throwing light upon their possible 
or probable mode of introduction. 
We have three species which Mr. Wollaston himself states to 
be usually winged elsewhere, but which are apterous in Madeira. 
These are Metabletus obscuroguttatus, Calathus fuscus, and Brady- 
cellus fulvus. I am inclined to believe that there are a few others 
which will come under this category, but it is very difficult to get 
information as to the winged or apterous character of particular 
species. These insects, however, have evidently become apterous 
since their introduction into Madeira. We have therefore no 
difficulty in accounting for their introduction, and, as no othcr 
change in their external characters has been effected, we may 
suppose it to have been comparatively recent. ' ; 
Next we have those genera which, though apterous in Madeira, 
are wholly or partially winged elsewhere. These comprise a large 
number of species, and are twenty-two in number, as follows :— 
Carabidze : Cymindis, Dromius, Metabletus, Scarites, Apotomus, 
Loricera, Leistus, Calathus, Olisthopus, Argutor, Cratognathus, 
Bradycellus, Trechus. Philhydrida: Hydrobius. Byrrhidee : 
Syncalypta. Curculionidee : Phleophagus, Tychius, Smicronyx. 
Heteromera: Phaleria, Helops. Staphylinidze: Homalota (1 
sp.), Othius, ere we are carried back to a remoter epoch for 
the introduction of the winged ancestors of the Madeiran species, 
since not only have the wings become aborted, but the insects 



themselves have become modified into distinct and often very 
well marked species. 
The next category consists of apterous genera which are peculiar 
to Madeira and the other Atlantic islands, but which are allied 
to winged groups, as follows : 
LElliptosoma.—Closely allied to Loricera, winged. 
Lurygnathus.—An abnormal form of Licinides, most of which 
are winged. 
Zargus.—An abnormal form of Chleniides, winged. 
Thalassophilus.—Allied to Trechus, winged. 
Tarphius.—Belonging to the Colydiidze, most of which, Mr. 
Pascoe informs me, have wings. 
Coptostethus.—Allied to Cryptohypnus, winged. 
Caulophilus.—Allied to Phleophagus, winged. 
Lipommata, Mesoxenus, Caulotrupis.—Anomalous genera of 
Cossonides, which are often winged. 
Acalles, Torneuma.—Aberrant genera of Cryptorhynchides, 
most of which are winged. 
LEchinosoma.—Doubtful affinities. 
Atlantis, Cyphoscelis, Laparocerus (Laparocerides).—A very 
isolated group. 
Anemophilus, Scoliocerus.—Allied to Trachyphloeides, some of 
which are winged. 
Lichenophagus.—Allied to Cenoposis and Omias, some of 
which are winged. 
Xenorchestes.—Allied to Choragus, winged. 
LEllipsodes.—Closely allied to Cryfticus, some of which are 
winged. 
Hadrus.—Belongs to an apterous group of Opatrides, many of 
which are winged. 
Macrostethus.—Belongs to Ccelometopides, all of which are 
apterous, but comes next to the Zenebrionides vrais, of La- 
cordaire, which are mostly winged. 
Xenomma.—Belongs to the Aleocharides, which are winged. 
Mecognathus.—Allied to Sunius, winged. 
Metopsia.—Allied to Phlaobium, winged. 
Here we have indications of an introduction of forms at a 
still more remote epoch. In many cases the modifications of 
structure have been so great as to produce distinct generic forms, 
while these remain still allied to winged European genera. In 
other cases, however, the modifications are still greater, and the 
affinities are with groups which in Europe are wholly apterous. 
Such cases as Hadrus and Macrostethus, which belong to small 
groups of wholly apterous genera, are difficulties on the theory 
of transmission over the sea. But two considerations render 
this difficulty less real than apparent. They all carry us back to 
a very remote epoch; and, knowing what we do of the in- 
stability of the apterous condition, we may fairly conclude that 
the groups in question were, at that time, in a partially winged 
state. At or near this same remote epoch, the Madeiran group, 
as indicated by the submarine bank now connecting the several 
islands, probably formed one more extensive island, and the 
distance of ocean to be traversed would then have been con- 
siderably less than it is now. 
If the various groups of facts which I have here set forth, 
respecting the distribution of apterous and winged species and 
genera, are fairly considered as a whole, I think they will be seen 
to be quite inconsistent with the theory of that distribution having 
been effected by a former land connection with Europe ; and, 
considering that we are necessarily ignorant of many of the ways 
by which organisms are transmitted across ocean barriers, such 
transmission seems to be indicated in the case of the Madeiran 
Coleoptera, not by means of drift wood and ocean currents, 
which Mr. Murray thinks must be the most efficient means of 
transport, but by some mode in which their wings are called into 
play, which can only be by a passage through the air when 
assisted by gales and hurricanes. 
There is one other group of islands which seems well adapted 
to offer a crucial test of the correctness of the theory of land- 
connection. The Azores are more than twice as far from Europe 
as the Madeiras, and, what is of still more importance, they are 
cut off from it as well as from the Madeiras by a broad belt of 
ocean of the enormous depth of nearly 15,000 feet. We may 
feel pretty confident, therefore, that if both groups have once 
been united to the continent, the separation of the Azores is by far 
the more ancient event ; and any theory which requires the Azores 
to be the most recently separated must be strongly supported by 
independent evidence to render such an improbable supposition 
acceptable. If the Azores date the origin of their insect popu- 
lation from a remote epoch when they were connected with 

