are destroyed at temperatures at which careless workers frequently pour their agar plates, while others refuse to grow at ordinary temperatures or even at blood heat, grow best at 50°-60°C., and are not killed until the temperature exceeds 70° or even 75°C. Finally, a race of Bacterium anthracis incapable of producing spores has been developed by growing the organism in media containing phenol; another non-virulent race bearing swollen, terminal spores, "drumsticks," by growing the organism in compressed air; and still another race destitute of virulence by cultivating it at temperatures above 40°C. These are not exceptional cases, similar care being necessary in all directions if one would avoid erroneous conclusions.

Naturally, every successful experimenter will vary his culture media in all sorts of ways in order to learn as much as possible of the organism under consideration, but at the same time he will determine its behaviour on the standard media, and will keep a very careful record of all that he does. The bacteriologist should make it an invariable rule to repeat every experiment two or three times, at the very least, and generally after an interval of some months or years he should repeat all his experiments. Even then he will fall into errors enough. He certainly should proceed with as much care as the chemist, and in many directions the work passes naturally over into chemistry. If quantitative or volumetric analysis requires all sorts of precautions and excess of care to avoid errors, no less does this youngest of all the sciences.

A few words respecting the occurrence of bacteria in normal plant tissues will be in place before concluding these general remarks. It goes without saying that such minute and universally distributed bodies as bacteria are likely to be found at times almost anywhere, even in plant tissues which seem to be healthy, just as they may sometimes occur in the blood stream of healthy animals, but they are not normally present in the tissues of plants. All carefully conducted experiments have

15

led to this conclusion. The reader who wishes fuller information may consult papers by Laurent," Buchner, Lehmann," 14b (14) Sur la pretendue origine bacterienne de la diastase. roy. de Belgique, T. X., pp. 38-57.

Bull, de l'Acad.

15 (15) Notiz betreffend die Frage des Vorkommens von Bacterien in normalen Pflanzengewebe. Muench med. Wochenschrift., 1888, pp. 906–907.

16 (16) Erklarung in Betreff der Arbeit von Herrn Dr. Hugo Bernheim, etc. Ibid, 1889, p. 110.

Fernbach" Vestea,18 Kramer,19 and Russell.20 Even when purposely introduced into living tissues they refuse to grow or spread but little and finally die out," unless they possess specific pathogenic power in which case the result is quite different.

The diseases which will be discussed in the following pages may be divided into four classes:

(1). Diseases of clearly established bacterial origin.

(2). Diseases which appear to be constantly associated with bacteria and which are probably due to some specific organism, but full proof of which has not been furnished.

(3). Diseases said to be more or less closely associated with the presence of bacteria and ascribed thereto, but in which little or no proof has been brought forward to establish the causal relation.

(4). Communicable diseases which have been ascribed to bacteria but associated with which no organism has been found and which are probably of non-bacterial nature.

On the whole it would perhaps be more logical to divide the following pages into four chapters in the way I have specified, but for practical reasons it has seemed better to discuss all of the diseases of a given plant in one place. I have, therefore, arranged the material by hosts, but will at the close try to summarize the whole subject in the manner above indicated.

It will certainly be some time, probably many years, before we have anything like a permanent scheme of classification for the bacteria. Our knowledge is still too incomplete. Meanwhile, we have to do the best we can with the present systems, all of 11 (17) De l'absence des microbes dans les tissus vegetaux. Annales de l'Inst. Pasteur, 1888, pp. 567-570.

18 (18) De l'absence des microbes dans les tissus. Ibid., 1888, p. 670–671. 19 (19) Bakteriologische Untersuchungen ueber die Nassfäule der Kartoffelknollen. Osterreichisches landw. Centralb. I, Heft 1, 1891.

20 1. c.

21 Lominsky: (20) On the parasitism when introduced into plants of some disease-causing microbes (Russian). Wratch., 1890. No. 6, pp. 133-135. Russell: 1. c.

Kornauth: (21) Ueber das Verhalten pathogener Bakterien in lebenden Pflanzengeweben. Centrb. f. Bakt., Parasiten-Kunde, u. Infectionsk, I Abt., Bd. XIX, No. 21, 8 Juni, 1896, pp. 801-805,

which are more or less arbitrary and unsatisfactory, and all of which are liable to be set aside at any time. I have here adopted Migula's system" which seems to me very convenient, and on the whole the most satisfactory of any that has yet appeared.

Before proceeding to the body of this review it only remains to say that every effort has been made to deal impartially with the material in hand, and to present the essential ideas of the writers as concisely and accurately as possible. To this end the original papers have been consulted in every instance, unless otherwise stated in the text. So much vexation over wrong references has been experienced in time past by the writer that he has himself been at special pains to give full and accurate citations. It is to be hoped, therefore, that the reader will have no difficulty in finding the original papers. An endeavor has also been made to bring the subject fully up to date but it is quite likely that some worthy papers may have been overlooked, owing to the many languages and the ever increasing number of places of publication.

THE MEANING AND STRUCTURE OF THE
SO-CALLED "MUSHROOM BODIES" OF

THE HEXAPOD BRAIN.

BY F. C. KENYON, PH. D.1

In looking at a series of sections of the brain of a hexapod, especially of a hymenopteron, the most notable structures are two pairs, one to each side, of large cup-shaped bodies of "Punkt substanz," or, what in the light of our present knowledge of nerve structure is better denominated fibrillar substance. Each of these cups is filled to overflowing with cells having large nuclei and very little cytoplasm. From the under surface

:

22 Migula Schizophyta : (22) Schizomycetes. Die Natuerlichen Pflanzenfamilien (Engler u. Prantl). I Teil. 1 Abt. a, Lief. 129 8vo. p. 44, Leipzig, 1896. This is the forerunner of a larger work soon to be published by Gustav Fischer, Jena.

1 1 Clark University, Mass.

each of these cups or "Becher" there descends into the general fibrillar substance of the brain a column of fibrillar substance which unites with its fellow of the same side to send a large branch obliquely downwards to the median line of the brain and an equally large or larger branch straight forwards to the anterior cerebellar surface. (Fig.

B.)

Long before our present methods of sectioning and staining had found general application in the study of animal structure, or as early as 1850, the French naturalist, Dujardin, discovered these bodies in transparent preparations in toto of the brains of certain Hymenoptera and Orthoptera. From their somewhat folded appearance he describes them as "lobes à convolutiones," and compared them with the convolutions of the human brain, and even thought them associated with hexapod intelligence. Fourteen years later, Leydig, using the same methods confirmed Dujardin's discovery in working with the brains of the ant, bee, and wasp, and described them as "gestielter Körper." In 1875 Rabl-Rück hard identified the bodies in Gryllus italicus, Locusta viridissima, and Dycticus verrucornis, and correctly described the form of the "cup" under the term "Rind Körper." The very next year ('76) Dietl's application of the section method to the subject confirmed and perfected previous descriptions, and, struck with the resemblance to mushrooms, he adopted the name of "Pilzhutförmiger Körper," a conception later used by Packard (mushroom bodies) and by Bellocici ('82) (corpo fungiformo).

As to the intellectual function of the bodies, not all of the early writers supported Dujardin's inference. They were supposed to be connected with sight; but Rabl-Rückhard showed that they are fully developed in a blind African ant (Typhlopone). Dietl was loth to acknowledge an intellectual function, even though he found the organs more highly developed in Hymenoptera than in Orthoptera. But Forel (74) adhered to Dujardin's supposition, and showed that among Hymenoptera even of the same species the bodies are most prominent where one usually recognizes most intelligence, as in the worker bees and ants, while they are small in the females and the males. Brandt ('76) two years later in a note on the brain of Hymen

optera makes the same observations as to the differences in the same species, while Berger ('78) considered the structures as organs of projection of the first order."

The supposition of Dujardin obtained its best support so far as the older methods would avail in the comprehensive work of Flögel ('78) covering the whole group of hexapods. Here, one may see at a glance that the development of the structures largely coincides with the development of intelligence, as shown by the following abridgement of his table: A. The four cups completely developed. 1. Very highly developed,

Vespa.

Apis, Formica Pompilus, Ichneumonidæ.

Blatta.

Cossus, Sphinx, Vanessa.

If such a superior neural function is indicated by the testimony and work of the earlier writers, it may well be asked whether recent neurological methods will bring out the structure of the hexapod brain as well as they have that of the other invertebrates and that of the vertebrates, and whether they will lend this view support. First, it may be noted that the physiological experiments of Binet ('94), which are those of