May 21, 1885 | 
NATURE 
53 
remained in the nest, and the few that were about seemed 
agitated and stung virulently. Probably the mass of them had 
been driven off or eaten by the woodpeckers. The tunnel the 
latter had made was about two inches in diameter and four 
inches long, bored horizontally in, and ending in an irregular- 
shaped egg-chamber about ten and a half inches in cross diameter, 
but narrowed by the branch of pyingado which pierced the nest 
through and through, and crossed the egg-chamber diagonally. 
The bottom of this chamber alone was smooth, but there was 
no lining, and the two translucent white eggs of the woodpecker 
had rested on the bare boards, so to speak, of the ants’ house. 
In the excavations ccc made by the ants themselves there were 
neither eggs, larvze, nor pupe; probably these all had been 
removed when the woodpeckers invaded the nest. 
CHARLES BINGHAM, 
Deputy Conservator of Forests, British Burmah 
Henzada, British Burmah, April 12 
Staminody of Petals 
The cases of staminody of petals not being very frequent, it 
may be of interest to draw the attention of the readers of 
Nature to such a modification as observed in Fuchsia. 
The places of the four petals of the flower examined are 
occupied by four almost colourless filaments of an average length 
of three-fifths of an inch. Each of them bears on its top a 
nearly circular dark red lamina of three-tenths of an inch dia- 
meter. These laminz are so strongly vaulted as to have the 
shape of a segment of a globe, the hollow side being turned out- 
ward, the convex inward. At the base of the lamina, z.e. at the 
top of the filament, a short protuberance is seen, resembling in 
external shape the lower part of an anther. This anther occupies 
the concave side of the lamina and is consequently turned out- 
ward. Though the anther of one of the petals is only slightly 
developed, yet it may be admitted as a matter of fact that, 
instead of petals, this flower has produced four stamens, whose 
anthers bear a petaloid appendage. A microscopic examination, 
namely, showed not only the peculiar composition of the anther- 
wall, but also the presence of pollen-grains. 
Of the stamens, properly so called, the outer whorl is present, 
but the inner one is only represented by two of the four. One 
of these two is inserted in the ordinary way, viz. at the base of 
the petal. The second, however, has grown together half way 
up with the petal’s filament ; there it has, in consequence of a 
spiral turning, arrived at the back side of the petal, whence it 
bends obliquely outward. By this union the impression is 
created of a stamen rising from the back of the (modified) 
petal, concealing its anther in the Jamina’s concavity. This 
occurrence brings to recollection the case of MJonarda fistulosa 
as cited by Maxwell T. Masters from Turpin (‘‘ Vegetable 
Teratology,” p. 298), with this difference, however, that what is 
probably only adhesion is mistaken for petalody, whilst the case 
above described offers an antheroid petal grown together with a 
true stamen. J. C. CosTERUS 
Amsterdam, May 4 
Catalogue of Fossil Mammalia in the British Museum, 
Part I. 
In the review of the above work in a late number of NATURE 
(vol. xxxi. p. 597) the reviewer entertains such a complete misap- 
prehension of my system of naming the premolar teeth of typical 
heterodont Eutherian mammals that I must~beg space to 
correct it. 
The reviewer asserts that this system is untrue because it im- 
plies that in general with a smaller number than the full comple- 
ment of four premolars the diminution must have commenced 
with the first, proceeded with the second, andso on. In reality it 
implies nothing of the kind, and if he had taken the trouble to 
turn to pp. 152 (No. 39,732) and 174 (No. 48,787) he would 
have seen instances where I have mentioned the absence of the 
middle teeth (.2 and fm.3) and the retention of the terminal 
teeth (fm.1 and fm.4). Similarly in the ‘ Palzontologia 
Indica,” ser. ro, vol. iii. p. 48, I have adopted the same system 
for the incisors, and have shown that in Hippopotamus it is 7.2, 
and not 7.3, that disappears in some species. 
IT am well aware that in many of the Insectivora and 
Chiroptera there is often great difficulty in deciding on the 
homology of the individual premolars when these are reduced in 
number ; and the reviewer might have noticed that in the former 
order I have not ventured to definitely determine the position of 
any tooth in advance of the last premolar. Among the Chiroptera 
I have considered the three premolars of Vesfertz/io (p. 13) as 
homologous with the last three of the typical series, as there is 
apparently no evidence to the contary; the small size of 
pm. 3 indicates, however, that an allied genus may retain only 
pm. 2 and pm. 4; but the minute size of the one tooth in advance 
of pm. 4 in Rhinolophus has induced me to regard it as pm. 3, 
although it may be fm, 2. 
The advantage of the system employed in the ‘‘ Catalogue” 
is well instanced when we contrast the premolar dentition of 
Canis, and Lepus or Theridomys ; the homology of the last 
tooth of this series (and there is only one in 7herzdomys) being 
at once seen, whereas it is entirely lost if we employ a method 
like that used in Dr. Dobson’s ‘‘ Catalogue of Chiroptera,” 
where the actual first tooth in each genus is called the first of 
the series. I claim for the system adopted by myself every ad- 
vantage in those cases where it is possible to determine the 
homology of the individual premolars in any form in which the 
number does not exceed four; and even in cases where such 
determination is not absolutely certain, the error can be but very 
slight, and does not lead to the utter confusion caused by the 
system (or, rather, the want of system) which I presume the 
reviewer would prefer. 
When we come to those mammals in which the number of 
premolars is more than four, my systen fails; and, in view of 
this, some German writers have adopted the plan of numbering 
the premolars the reverse way—z.e. terming the premolar next 
the first molar fm. 1, and then counting towards the incisors. 
Although this system would be advantageous if we could always 
be sure of the division between the premolars and molars in 
homceodont mammals ; yet it has several di-advantages, and has 
not, therefore, been adopted. 
In reference to the suggestion of your reviewer, that instead 
of making a catalogue of the fossil Mammalia in the collection of 
the British Museum (as I was instructed to do by the Museum 
Authorities), I should have made one of all the known species of 
fossil Mammalia, any person having the slightest pretence to any 
knowledge of the present state of mammalian paleontology 
would have at once known that it would be utterly useless to 
attempt any such work at the present time, when new species 
and genera are being made almost daily, and a host of those 
already made are as yet but empty names. 
As a minor matter, I may mention in regard to the lower jaws 
of Crossopus, alluded to in the review, that their identification 
rests solely on the authority of Prof. Sir R. Owen, and that per- 
haps I have acted in a too conservative spirit in admitting them. 
Harpenden Lodge, May 2 RICHARD LYDEKKER 
Fossil Insects 
“« THE Earliest Winged Insects of America ; a Re-examination 
of the Devonian Insects of New Brunswick in the Light of 
Criticisms and of New Studies of other Paleozoic Types,” is 
the title of a brochure by Mr. S. H. Scudder, of Cambridge, 
Mass.. recently published. 
These Devonian insects are fragments of five wings ; a sixth is 
now dropped, as ‘‘ too imperfect for any satisfactory discussion, ” 
though in 188r its description filled about two quarto pages. 
These insects have been, since 1865, so often discussed that their 
literature is a rather voluminous one. A number of far-reaching 
conclusions elaborated by the author would have to be aban- 
doned if the determination of the insects should be proved incor- 
rect. This I endeavoured to do in Bull, Mus. Comp. Zool., 
viii. No. 14, Cambridge, 1881, and in NATURE, xxiii p. 483. 
The principal aim of the author’s new paper is to show that my 
determinations are erroneous. Concerning his statement that I 
haye studied in nature only the (in most cases poorer) reverses, 
I may remark that his paper gives nothing more, after his study 
of the obverses ; even less for Gerephemera. 
These Devonian insects have been decidedly unfortunate from 
the very outset. Eminent paleontologists denied their Devonian 
origin, and put them to the Carboniferous or to the ‘‘ Ursa Stufe ’ 
of the sub-Carboniferous. One of the insects, Xenoneuraz antiqu- 
orum, said to possess a stridulating organ on the wing, caused an 
unusual sensation. Poetic paleontologists were delighted to be 
introduced by this insect to the sounds of the Devonian woods. 
Now these woods are silent again, except in some text-books. 
“©Tt does not appear reasonable,” said the author, “‘to maintain 
