3<D4 DISCOVERY REPORTS 



formalin during a three-months' visit that I made in the spring of 1955 to Arrecife, Canary Islands. 

 During part of mis time I was fortunate enough to have the co-operation of Dr G. O. Mackie. 



After having worked on these specimens, which range in size from 11 to 180 mm. in float-length, 

 for two and a half years at the British Museum (Natural History), I heard that Miss Eleanor Dodge 

 was preparing to work at Miami, Florida on the toxin of Physalia. I wrote to ask if she would be kind 

 enough to secure me some larvae should they be available. On 14 January 1958 I received from her 

 a most welcome series of young specimens measuring from 4 to 15 mm. in length. Miss Dodge was 

 also kind enough to send me from Crandon Park Beach, Biscayne Key, off Miami, a second valuable 

 collection of younger larvae measuring from 1-2 to 5 mm. in length, collected on 20 January 1958, 

 which I received on 14 February 1958. I was thus able to figure and insert into the typescript, which 

 I had already completed, information on those stages of development at which budding begins. In 

 this way I have been able to complete the picture of the origin and pattern of budding in Physalia. 



It is almost impossible to examine specimens fully until the air in the float has been displaced by 

 fluid, or the float cut off. My most rewarding work has been done by combining successive dissections 

 with photography. The methods employed had to be purely exploratory to begin with, because 

 previous accounts of the arrangement of the groups of buds in Physalia were too superficial and 

 illustrated either not at all (Schneider, 1896) or so inadequately (Steche, 1910; Okada, 1932, 1935) 

 that they could not be checked and built upon. In fact it was necessary to start de novo. Now that 

 some kind of pattern has been recognized it may be thought that the exploratory phase of the work was 

 unduly long. The results must be published now before further refinements of the analysis can be 

 made. But it is hoped that someone in the future will be able more quickly to examine a series of 

 specimens (comparable with the young specimens to which I devoted much attention), and will find 

 out to what extent there is variation of the pattern observed and illustrated in this memoir. 



It is regretted that a better photographic technique was not available, but it was not possible to make 

 repeated dissections in the photographic studio of the Museum. Also, the carrying of such dissections 

 in dishes of formalin over long distances from laboratory to studio was found to cause too much 

 disarray of the specimens through swirling motions of the liquid. Recourse was therefore made to 

 dissection on the stage of a Bausch and Lomb binocular dissecting microscope set up on the base of 

 an old Leica photocopier. The arm carrying the viewer, a Leica camera and lamps was swung to one 

 side. When the dissection had been displayed, the body of the microscope was slid out and camera and 

 viewer were moved into position. 



- A method of dissection that I would recommend to anyone wishing to examine the structure of 

 Physalia for the first time is as follows: First remove the float, leaving the primary polyp and appen- 

 dages of both oral and main zone attached to a narrow strip of the bladder-wall. Secondly, sever the 

 oral zone at the basal internode from the main zone. Thirdly, sever the first two cormidia by a cut with 

 fine scissors just to the oral side of the main tentacle. Then make a similar cut a little to the oral side 

 of the chief tentacle of cormidium V. This will leave the three most representative cormidia attached 

 to a strip of the bladder-wall, which can be turned about and folded to reveal the internodes between 

 the three cormidia. A well-relaxed specimen about 4-5 in. in float-length would probably not be too 

 complexly branched for recognition of the basic pattern of arrangement of branches, polyps, tentacles, 

 and gonodendra. This basic pattern is described on page 328. 



The structure of contracted specimens is exceedingly difficult, if not impossible, to analyse satis- 

 factorily, so live specimens to be examined in this way should be carefully anaesthetized in magnesium 

 chloride (7^% in fresh water) and fixed in formalin. It was found that expansion was aided by 

 injecting with a fine needle both the magnesium chloride and the formalin into the gastrovascular 

 space near the base of the chief tentacle. The muscular walls were found to be self-sealing. Air can 



