Search

Causes of rich and poor Year-classes of Herring. This paper continues RUNNSTRØMS investigation on herring larvæ during 1932-35 and is based on material obtained by hatching artificially fertilized eggs as well as on plankton material from the West Coast during late winters. In a preliminary account of the work (1940) the author pointed out that the artificially bred larvæ as well as those found in the sea went through a "critical stage" when the yolk was consumed, and that the greater number of the larvæ were unable to survive this dangerous epoch. This stage was described in detail and the cause of the great mortality ascribed to lack of suitable food. It was furthermore pointed out from experience gained by tow-netting that the herring larvæ were not encountered at the surface in sunshine but only at some depth. By measuring many hundreds of larvæ it was found that the critical stage occurred about the end of March. In the present paper the experiments with fertilizing and hatching of the eggs and feeding of the larvæ are treated in more detail. The salinity was found to be unimportant within the range 22,17 -33,57 per mille. The influence of temperature investigated more closely and it was found that the biological zero-point of the herring egg is situated at approx. ÷ 1.2° C, and that the product of the time and the average temperature (counted from the biol. 0-point) was nearly constant in all hatching experiments, the latest as well as those made in earlier years. It was furthermore found that light is of great importance for the act of feeding of the larvæ. Contrary to earlier conceptions it was found that a vegetarian diet alone was unable to keep the larvæ alive in early life. It was observed that the larvæ, provided suitable illumination, were catching and eating barnacle-nauplii (Balanus balanoides) and those of the "brine shrimp" (Artemia salina) even before the yolk was completely used up. Such food enable the larvæ to be kept a long time beyond the critical stage. Tow-nettings in the later years corroborate the findings mentioned above, that the larvæ avoid direct sunshine and keep to a twilight zone. In the period succeeding the great reduction in the number of larvæ found in the plankton, great quantities of dead and semidissolved herring larvæ appeared in the catches and in these place the sea was exceptionally poor in nauplii. In other catches it appeared that eggs and nauplii of copepods (above all of Calanus finmarchicus) most commonly are found as the first gut-content of the herring larvæ, along with remains of diatoms. The later spreading and transport of the herring fry takes place both northwards along the coast with the coastal stream and into the fjords by means of the tidal currents. Investigations in the little landlocked fjord Nordåsvannet (near Bergen) from spring to autumn 1941 showed the occurrence of herring fry but only after it had passed through its metamorphosis into "whitebait" which must have been carried into the fjord by the very strong tidal current through the narrow entrance.

In the first half of this paper is given an abbreviated account of the biology of the spur dog. In the latter half of the paper the Norwegian dogfish taggings are dealt with: a. A short description of the tags used. b. Short accounts of the different tagging cruises. c. The general migration route seems to be betvveen the Shetland-Orkneys (summer) and the Norwegian west coast (winter). Recaptures from the Shetland taggings are recorded also in the Vestfjord (arctic circle) and in the Barents Sea (Thor Iversen Bank). d. The yearly fishing mortality runs about 7 %· e. The natural mortality appears to be about 20 % per year. f. The stock on the Norwegian fishing grounds (Jan.-March 1959 & 1960) is estimated to roughly 100 000 tons. g. Some peculiarities in the recapture pattern this summer and autum (1960) indicate that a substantial portion of the stock has remained in Norwegian coastal waters throughout the year.

Nesten alle som anvender en objektiv bedømmelsesmetode som støtte for organoleptiske iakttagelser av fisks ferskhet har valgt bestemmelse av flyktig N. Dette er et samlebegrep for ammoniakk og metylderivatene av denne. En av dem: trimetylamin, synes særlig egnet for øyemedet da mengden av den stiger svært langsomt så lenge fisken er god menneskeføde og deretter tiltar i stadig økende målestokk. Man får en glatt og karakteristisk kurve for trimetylamin, mens ammoniakkkurven forløper på en måte som gjør den lite egnet. De øvrige, mono og dimetylamin er kvantitativt sett av mindre interesse, om enn dimetylamin tillegges en viss betydning. ENGLISH In all countries interested in the transport and storage of fresh fish and processed fish products an objective method of controlling the freshness of the fish, viz. the estimated content of volatile N or one of its components, especially trimethylamine N, has now generally been adopted. Owing to different methods of sampling and performing analyses, comparison between the results of different investigators can not be easily drawn. We therefore decided to make some research concerning the volatile forms of nitrogen appearing as decomposition products in the bacterial breakdown of fish, and with special respect to analytical procedure. The results of this work are laid down in this paper.

The material dealt with in this paper contains 975 specimens of fresh-water eels from the Norwegian Skagerrack Coast. Of these specimens 674 (9 males, 665 females) were yellow eels collected in the sea near the shore; the rest, 301 specimens, were taken in lakes and rivers near the town of Arendal. Of the latter group 150 specimens (35 males, 115 females) were silver eels, 48 specimens (females only) were yellow eels, and 103 specimens were young ones of indeterminable sex. Attention is especially drawn to the two different types of yellow eels, viz. broadheaded and narrowheaded ones. The measurements (external head proportions in relation to total length and thickness of the body) show a marked difference between the extreme specimens of the two types, but also that they are well combined by intermediate forms. Cranial measurements of young eels show the same variation as the external head measurements in larger eels. By counting the vertebrae in typical specimens no difference in this character was found between the two types. By changing from yellow eel to silver eel the body increases very much in thickness; measurements show that this fact is sufficient to account for the relatively narrower head in the silver eel. The size of the eyes by changing from yellow eel to silver eel increases by 62 per cent in the males, 39 per cent in the females. The analysis of the stomach content of the eels from the sea shows that crustaceans, and partly small fishes, are the most important food animals. To a certain degree the broadheaded eels prefer fish, the narrowheaded ones crustaceans. In fresh water insect-larvae of different kinds together with small snails constitute the main food of the eels.

This paper, by way of introduction, presents a very brief review of the known, pronounced and large sulphur-deficient areas of the world. The major factors determining the sulphur supply of cultivated plants are discussed briefly on the basis of literature studies. The importance of the atmosphere as a source of sulphur has been given more attention than the other factors. Results of experiments and other investigations in this field, as well as the opinions held on the sulphur fertilizer question in the Nordic countries, are also treated somewhat more in detail. The list of references comprises a very limited selection from the comprehensive literature on the subject in question.

Since 1959 the Norwegian sprat investigations have been considerably intensified. In this connection extensive hydrographical investigations have also been carried out. In this paper an account is given of the distribution and movement of the different water masses in the area shown in Fig. 1. The distribution of sprat eggs and larvae in relation to the hydrographical conditions has also been discussed. A warm surface layer with high content of oxygen, and a low phosphate content was separated from the underlying water masses by a transition layer in which the stability was particularly high. A core of Atlantic water with relatively high salinity, temperature and oxygen content, and a low content of phosphate, penetrated into the Skagerak along the edge of the North Sea Plateau, continuing along the Swedish and Norwegian coasts, constantly losing its original character (Fig.s 2 and 3). Current measurements were made at an anchor station (St. 410 on Fig.s G and 7). During the summers of 1960 and 1961 drift-bottle-experiments were carried out. The assumed drift is shown in Fig.s 8 and 9. The number of days drift is indicated at the point of recovery. The investigations indicate the existence of a cyclonic vortex in Skagerak. Persuming dynamic equilibrium this vortex is in good accordance with the depth of the boundary surface between the upper layers and the deep-water masses found by Dietrich ( 1950). Sprat eggs and larvae were collected in vertical hauls with a one-meter egg net. Examples of their distribution in the area are given in Fig.s 10 and 11. During the last 10 years sprat larvae have been caught in much smaller numbers outside western Norway than one might expect in relation to the concentration of eggs and larvae in Skagerak. The present author is of the opinion that because of the cyclonic vortex mentioned above, the larvae are forced to remain in the Skagerak for a relatively long period. When they finally are carried to the area off western Norway, they have grown large enough to avoid the plankton nets used.

Der er gitt en oversikt over tidligere undersøkelser av fisk og dens holdbarhet under lagring, over fiskens sammensetning og dens bakteriologi, fiskemuskulaturens oppløselighetsforhold i vann og de prosesser som foregår i muskulaturen når fisken blir drept og etter at den er drept. Der er angitt kjemiske, bakteriologiske og biokjemiske analysemetoder som anvendes under studier av lagret fisk i dette laboratorium. Der er foretatt undersøkelser over: l. framstilling av jernsera, 2. eggehvitestoffenes oppløselighet i avhengighet av jonestyrken, 3. aminosyrer etter SØRENSENS formolmetode, 4. CONWAY og BYRNES bestemmelsesmetode for flyktig Nidet den er prøvet foruten for ammoniakk også for trimetylamin, 5. destillasjon av samme flyktige baser ved vanlig trykk, 6. innhold av forskjellige kjemiske stoffer i fersk og lagret fisk i form av farse, pressaft, ekstrakt og serum, 7. bakterieinnhold i fersk og lagret fisk, 8. metylenblåttavfarging, 9. indolbestemmelser i fersk og lagret fisk og tilslutt 10. er der utført en rekke lagringsforsøk av sløyet råfisk og fiskefarse hvori er bestemt totale flyktige baser, ammoniakk og trimetylamin. Der ble funnet følgende re sultater: l. Sera framstilles av mager fisk under tilsetning av 30 ml kolloidalt jernhydroksyd pr. 100 g farse. 2. Jo mindre farsen fortynnes med vann desto relativt mer eggehvite går i oppløsning i ekstraktet. Mengden av de andre stoffer i ekstraktet er proporsjonal med farsekonsentrasjonen innen vide grenser. 3. Aminosyrebestemmelser kan bare utføres i sera, men opptil 80 mg eggehvite pr. 20 ml serum influerer ikke på resultatet. 4. CONWAY og BYRNE metodikken som også anvendes av BEATTY og GIBBONS (1937) gir for mager fisk samme resultater som en får ved destillasjon ved vanlig trykk av sera. For sild faller trimetylamintallene noe høyere ut. Metodens fordel er å være meget tidsbesparende. 5. Destillasjon ved vanlig trykk gir en kontinuerlig avspaltning av flyktig N fortrinsvis fra eggehvite men også fra andre stoffer i farse, ekstrakt og serum. Avspaltningen av N fra sera av mager fisk er imidlertid meget liten. Analyseresultatene ved destillasjon av farse og ekstrakt faller derfor stadig høyere og høyere ut jo lenger en destillerer og kan tildels bli meget misvisende. Vi finner derfor at en kun kan bruke sera ved disse bestemmelser. Vi kommer senere tilbake til event. anvendelse av farse til samme øyemed under bruk av vakuum og lav temperatur.6. Når ekstraktet framstilles av 250 g farse i 1000 ml, ligger det totale N i ekstrakttørrstoffet mellom 13-16,3 g/100 g. Aminosyreinnholdet i torsk ligger mellom 69-98 mg/100 g. Total flyktig N, ammoniakk N og trimetylamin N øker i fisk som lagres, særlig ved høy temperatur, og dette forhold kvalifiserer disse stoffer som ledestoffer ved bestemmelse av fiskens friskhetstilstand til en hver tid. Fra våre hittidige undersøkelser framgår at bestemmelse av ett av disse stoffer, trimetylamin, er særlig egnet for øyemedet. 7. Det aerobe bakterieantall på fisken karakteriserer ikke fiskens tilstand helt pålitelig til en hver tid under lagringen. Det er ikke alltid de samme bakteriesamfunn som gror på fisken, hvorfor der vil inntreffe visse forandringer i vekstkurvene da de forskjellige bakterier har forskjellige veksthastigheter. Dessuten er startantallet av bakterier på den fisk som legges til lagring avgjørende for hvor hurtig fisken ødelegges. Dyrkning av spesielle bakteriegrupper ville langt sikrere gi et bilde av fiskens tilstand. 8. Metylenblåttavfarging som er meget bra ved melkeundersøkelser, har i sin noværende metodikkform på grunn av den lange avfargingstid når fisken enn o må betegnes som brukbar, ikke gitt anvendbare data. 9. In do l kunde på vises i torsk etter 5 døgns lagring ved 0° C. Senere økte mengden stadig. Den kvantitative bestemmelse egner seg som et kriterium på fiskens tilstand når man kan påvise indoldannende bakterier på den. 10. Lagringsforsøkene, hvorunder ble bestemt total flyktig N, ammoniakk N og trimetylamin N etter Conway og Byrne, viste at total flyktig N og ammoniakk-mengden først avtok for så å stige. Særlig kraftig skjedde stigningen etter det 10.-11. døgn. Tilbakegangen i ammoniakkmengden som også er påvist av SHEWAN (1938), skyldes antagelig at ammoniakken nyttiggjøres av bakterier som lett assimiler bar kvelstoffnæring. Trimetylaminmengden øker derimot helt fra begynnelsen av, men økningen går meget langsomt inntil det 9.-10. døgn. Fra dette tidspunkt av tiltar trimetylaminmengden sterkt og når raskt opp i høie verdier. Dette knekkpunkt på kurven ved det 9.-10. døgn, faller temmelig nøie sammen med de organoleptiske prøver vi foretok. Etter 9. døgn kunde vi fremdeles anvende fisken, men det 11. døgn var fisken helt skjemt, og det 12. døgn måtte den b8tegnes som råtten. Trimetylamininnholdet i lagret fisk er derfor et utmerket kriterium på fiskens friskhetstilstand og da bestemmelsen av dette stoff er enkel og kan utføres med stor presisjon, er det meget som taler for å anvende bestemmelsen herav til kontroll av råfiskens friskhetstilstand. ENGLISH: This paper is intended partly to be an introduction to the biochemistry and bacteriology of fish with special reference to advanced spoilage during storage and partly to present various experiments made in order to decide some doubtfull points in analysis and biochemistry of fish. The processes in the muscles of the dead fish immediately after death and later on are largely of a chemical and bacteriological nature and it would be wasted effort to try to understand the complicated reactions unless the best methods were to be used for their analysis and interpretation. Comparison can not be drawn with much of previous work, based on different methods of sampling and different methods of analysis. Minced muscles, press juice made at different pressures, water extract prepared at different temperatures, shaking times and concentrations, buffer solution extracts at different pH and ionic strengths and sera prepared by means of precipitants of various kinds have been used as samples for the chemical analyses. The analytical methods are also of different kinds and very often the investigators have forgotten to compare their new methods or their modifications of the old ones to the most suitable of those known. Determination of volatile N is now thought to be the best method in detecting incipient spoilage in fish and in following the sequence of breakdown of the muscle. Our attention has therefore especially been confined to these substances. As ammonia N is produced particularly from proteins during prolonged destillation at ordinary pressure due to presence of salts and high pH values at the boiling temperature, we made attempts to use water extracts and sera of the tissue as samples. Extracts, like minced muscles, failed to give constant values of volatile N, but sera prepared by adding 30 cc colloidal iron hydroxide per 100 g of minced muscle gave values of volatile N, which after a destilling time of 35--40 minutes remained practically constant. The time-consuming nature of the destillation method outlined made it desirable to use a micromethod which we found in CONWAY and BYRNES method for detern1ination of small amounts of ammonia N. Simultaneously with BEATTY and GIBBONS (1937), who used the method in fish spoilage investigations, we made a modification of this method, also for trimethylamine N determinations. It is very rapid of manipulation and can easily be applied to press juice, water extract and sera. By comparing the two methods for coalfish we found dose agreement, but for herring the Conway and Byrne method showed some higher value in trimethylamine N. The non protein fraction of fresh fish muscle is proportional to the concentration of the extracts, this not being the case with the protein fraction thereof, owing to the relative greater solubility of the proteins as the ionic strength of the extracts increases. The determination of the amino acid N after the SØRENSEN method is not influenced by small amounts of protein in not completely coagulated sera(= 80 mg in 20 cc). The amino acid N of cod muscle is 68-98 mg/100 g. The determination of the number of aerobic bacteria only is not absolutely reliable for characterisation of the condition of neither a fresh nor a stored fish as the sequence of the growth curve of the aerobic bacteria does not follow the same sequence as the chemical products formed during the storage. Fresh fish, highly contaminated from unclean handling is broken down much sooner than fish protected against such contamination. The same bacteria societies do not always multiply on the stored fish. The bacterial growth curves from different fishbatches are not identical. The bacteria which spoil the fish are also of a facultative and of an obligat anaerobic nature and especially the facultative anaerobic species or the reducing achromobacter are very important because the y produce trimethylamin. It is highly desirable to improve the bacteriological methods of testing fish. The methylenblue reduction method, so valuable and so rapidly performed with milk analyses, is somewhat troublesome in its present form in view of the long reduction time of extracts of fresh and still sound fish. It is possible to show the presense of small amounts of indol many days before we should class the fish as almost spoiled. Later on this content increases more or less, due to the presence of various indole producing bacteria. Storage experiments with whole gutted cod and minced cod muscle at 0° C show a slow increase of trimetylamine till about the 9.-10. day. From the 10.-11. day there develop distinct signs of incipient spoilage, as the amount of trimetylamine increases very rapidly. Volatile N and ammonia decreases during the first 4-5 days and then slowly increases till about the 8.-9. day. From the 10.-11. day the increase is very great. As to the decrease of ammonia, it might be probable that it serves to begin with as an easily available source of nitrogen for the bacteria. The organoleptic examinations on the 9. day proved the fish to be still sound, on the 11. day it was definitely stale, and on the 12. day it had become obviously spoiled. The content of trimethylamine of fish is therefore very useful as an indication of its freshness during storage. As the determination is very easy and may be performed with great precision, much speaks for its applicability in testing stored fish until a more adequate method can be found.

An echo sounder is adjusted for constant sound velocity, usually 1500 m/sec. Since the velocity of sound in the sea is varying with temperature, salinity and depth, a correction is necessary if accurate soundings are required. The present paper shows a practical method for the computation of this correction. 49 of the "G. Sars" stations, taken in the years 1951-55, are fairly well covering the Norwegian Sea (Fig. 3). These stations have been used as basic material for the computation of the depth correction presented in Figs. 4-10. The correction is given as percentage of measured echo depth for the following depths (meters): 500, 1000, 1500, 2000, 2500, 3000, and 3500.

During the last 10 years we have been experimenting in the laboratory at the Flødevigen Sea-Fish Hatchery with the rearing of several species of salt-water fishes. The intention has been to bring about better methods for experimental purposes and to get some information as to what factors may be of importance for the survival of the eggs and larvae in nature. In this paper we are dealing with herring, cod, and plaice, especially the results from the two last seasons. In order to attain success when rearing fish-larvae, especially those species having an air-bladder, it is important to control the gas content of the sea-water. It is essential that when larvae are reared in small aquaria at a pressure of but one atmosphere the gas content of the seawater must be kept relatively low. It is assumed that the gas pressure and not the content of gas is the chief factor. Larvae living in the sea at a water pressure of e.g. 10 metres will probably sustain a gas content in the sea-water which will be detrimental near the surface. In this connection it is worth mentioning that the young cod-larvae react at once to changes in water pressure. When the water pressure is about 5 metres the larvae in a closed aquarium will keep to the surface. When the water pressure is shut off the larvae will straightaway sink apathetic to the bottom, but swim up again when the water pressure once more is put on. Experiments indicate that cod-larvae with a big air-bladder may recover when they are transferred to an aquarium covered with silk net and lowered to a depth of 5 to 10 metres. The eggs of the pelagic spring spawners are not affected by reasonable variations in the gas content. If the gas content is not reduced in the aquaria the larvae will get gas-bubbles in the intestines. This does not occur till the yolk-sack is resorbed - and the intestines and gills are functioning. It is assumed that the gas is liberated when the sea-water passes the gills. Small gasbubbles are formed inside and swallowed. In the case of herring the gas-bubbles are fatal. In the plaice the gas-bubbles will pass through the anus - and are not likely to harm the young. In the case of the cod we have not observed the gas to be within the intestines. In early stages - about a week after the resorbtion of the yolk-sack - the gas occurs free in the cavity. The intestines may be ruptured-or the gas may originate from the early air-bladder. In older stages the air-bladder will expand, the young will turn belly upwards and die, or the air-bladder may be ruptured. In older stages of the plaice - up to a length of some centimetres at least - fine gasbubbles rnay accumulate in the veins, and the fish will succumb. In the laboratory the gas content may be reduced by filtering the sea-water slowly through fine sand. Great quantities of gas will accumulate in the sand. The composition of this gas O²/N² is somewhat the same as in the atmosphere, the content of oxygen being slightly higher. Some of the gas in the sea-water in the laboratory may originate from the atmosphere - by leakage in the pump bearings. Studies of the gas of the air-bladder of cod (by Miss Eva Henly) indicate that oxygen is the operative factor. The results of the rearing experiments have improved to a very high degree after the gas problem was taken into consideration. The larvae of herring, cod, and plaice (and other flat-fishes) will thrive for weeks in the small aquaria used, containing but 35 litres. The flat-fishes are ordinarily reared without any mortality to the bottom stage in numbers from 1 to 3 thousands in the same aquarium. In this way the critical period - the moment when feeding begins - has been overcome. Experiments have shown that the cod-larvae may live in filtered water - without food - for several days after the yolk-sack is resorbed, and still thrive when fed. If we want to rear the fish further we encounter other problems. Will the feeding with the nauplii of Artemia be sufficient in the long run? Do they contain enough vitamins? Experiments have been made to feed the nauplii with flagellate cultures for some days before giving them to the young fish. Those experiments are to be continued. Experiments to furnish the young herring, about 2 months of age, with freshly caught plankton were negative. The herring did not take the copepods or Daphnia, and the plankton introduced seemed to kill the Artemia. We are now using fine chopped Mytilus edulis - even for the herring. The volume must be augmented - larger aquaria, more water, and larger filters. Experiments to rear cod and plaice in our large rearing boxes for lobster using but slightly filtered water were entirely negative. The results were the same when transferring several species of flatfishes after they had passed the stage of transformation. At present we have a good sample of herring 6 months of age, plaice ½ & 1½ year, and soles 5 years of age - all brought up from eggs in the laboratory. The last season we had some mortality of the young when 2-3 months of age. The symptoms in the plaice indicated an attack of Vibrio anguillarum. In the herring the symptoms indicated Lentospora cerebralis. The cod died before we were aware of the cause of the mortality and could take steps to reduce it. The attack on the plaice was stopped by adding some dissolved salt to the sea-water used. The herring reared in a larger aquarium in the laboratory survived, but in an ordinary aquarium (35 l) they succumbed. The infection this year came probably from our water reservoir. This was partly emptied in early winter, the algae and animals living along the walls died - and we were not able to have it cleaned out because of high snow and frost. Several experiments indicate that the young cod especially is very sensitive to products of metabolism. In one instance the water was drawn from a container where few cod of 1-2 years of age were living. The fry died. But when the older fish were removed, and the container cleaned a new batch thrived. It is impossible to say wether the mortality in the first experiment is to be ascribed to the chemical effect of the products of metabolism or to bacteria. Our experiences from last season give a warning that bacteria and products of metabolism may be a problem of higher order. Our earlier investigations in the inner Oslo-fjord, highly contaminated by sewage, have shown that the survival of the eggs and larvae of the spring spawners here is very poor. And in some places, where contaminated water occurs, even the grown up cod may suffer a great mortality, evidently caused by Vibrio anguillarum. The problem of rearing for experimental purposes has been our main object. Experiments on a large-scale rearing have to be postponed till we have aquired a relatively full knowledge as to the factors of main importance. Besides this work we have made some observations on the behaviour of the eggs. Many of the observations have to be confirmed by special experiments. In order to bring those points under consideration we will, however, mention some of them. The hatching and rearing has ordinarily been undertaken at an ordinary temperature of the sea-water during the season. In some experiments the water has been cooled or heated artificially. As we have no thermostats available for that purposes it has not been possible to keep the temperature quite constant. At a temperature of approximately 2° C the cod-eggs hatch normally. herring-eggs will also hatch at that temperature, but the larvae will have difficulties in escaping the eggs, and a heavy mortality occurs. Temperature-variations between 2° and 10° give high mortality to cod-eggs. In some rearing experiments the mortality among the young may be somewhat higher when the temperature is raising. The question of the relative gas content here comes into the problem. Fungus may be detrimental to the eggs. If they are heavily attacked, the eggs will sink. This occurs especially in foggy weather, and when the water has a relatively low salinity. The attack may be reduced by transferring the eggs for half an hour to sea-water to which has been added some salt. In some cases the cod-eggs have been infected by a flagellate living close inside the shell. Both these pests have been observed in nature - the last mentioned even in the middle of Skagerak. Cod-eggs hatched in circulating - but not renewed - water (constant salinity), gave a poor result. Bacteria may here be involved - or the products of metabolism. The influence of light has been studied more fully. The experiences from the hatching boxes indicate that the hatching of the cod-eggs gives the best results near the window. It is, however, difficult to say anything conclusive of the influence of the light in this case. In the boxes with less light there will also be a higher growth of fungus. Experiments with plaice-eggs indicate that normal daylight near a window - but shadowed for direct sun - gave poor results. And in experiments in the open air the eggs have died. On the other hand the plaice-eggs will hatch well in total darkness. It appears, however, that the plaice-eggs hatched in total darkness will give a higher percentage of larvae with abnormal pigmentation. The dorsal side is quite white, only the fins are pigmentated. When those plaice in the bottom stage, or later, are transferred to our salt-water pond under normal daylight and normal food for several months, no alteration takes place. Experiments also indicate that the plaice-eggs react with light. They may sink in the middle of the day and come to the surface again during the night.