Professor Wim Holleman's approach to this work was inspired by the philosophy developed by Rudolf Steiner known as anthroposophy (see, for example, Lissau, 1987). In 1933, whilst studying the composting of manure, Holleman obtained a copy of Steiner's [Biodynamic] "Agriculture Course". In it were given a few details on biological transmutations of chemical elements by the plants used in the preparations recommended for the treatment of manure and compost heaps. The lecture course was given to a select group of farmers and others wishing to improve the quality of agriculture, who were already familiar with many of the more esoteric aspects of anthroposophy. The published transcript is therefore not easy to follow, and many have and are still continuing to interpret and work with it [one such interpretation is by Soper, 1976].

Thus Holleman's approach to this subject, i.e. in the first two introductory sections of this review article, differed from a more conventional approach. There was a greater emphasis on the qualities of the subject under consideration, and also of the mind-set of the scientific observer.

The Ancient Greek scientist and philosopher Aristotle divided the nature of matter into two parts, into Materie Prima and Materie Signata, i.e. quality and quantity, or in other words, into idea and substance. An object only attains reality for a conscious observer when the idea, or concept of the object (quality), is combined with the observer's perception of the object (quantity). In medieval times the emphasis was on the spiritual representation - the qualities (taken in their broadest sense) - of the object and not the object itself. As Holleman indicates, in section 2.3, such a one-sided viewpoint needed to be overcome. This was begun during the reformation and finished with the materialistic culture of the 19th century. Though this was a necessary development to counter the one-sided medieval world-view, a completely materialistic world-view is, according to Steiner and others, just as one-sided. For an ability to fully realise the reality of an object, both are required. Thus it was, in the first two sections and the fifth of this article, that Holleman made an attempt to reconcile materialistic science with an earlier alchemistic, or spiritual, way of thinking. The importance of this was the essence of Steiner's "Boundaries of Natural Science" (1920) [reinterpreted by Thomas (?)] and also his "Philosophy of Freedom" (1894), from which it was derived. [For those wishing to further explore the fundamentals of the natural world from an anthroposophical/scientific viewpoint, I can strongly recommend Lehrs (1985).]

The underlying methodological approach taken by Holleman was based on a Goethean, phenomenological approach, as interpreted and developed by Steiner (1897). This is essentially an empirical approach, allowing the observations to speak for themselves. Holleman's work concentrated on the building up of observations such that the one led on to the next (see also Koepf and Jolly, 1978). Kervran (1972) - see section 4.3 - by way of contrast collected together a wide range of often circumstantial observations in order to support a theoretical mechanism for biological transmutations. For further, practical insight into the Goethian phenomenological approach see, for example, Bockemuhl (1986) and Edelglass et al. (1992). In the Netherlands research based on such an approach is conducted by the Louis Bolk Institute, Driebergen.

The outer, material, directly perceivable nature of an object or organism under study (in this case chemical transmutations under the influence of Chlorella vulgaris) may be observed by a series of carefully conducted, essentially conventional experiments. In a letter to Mr De Vries of Leiden (unpublished, 1982), Holleman revealed that for him, what "...is perhaps the most important for this whole research: the mental transformation that we ourselves as scientists must make, should we wish to penetrate the processes of living nature with truth. This seems illusionary. We know almost nothing of the reality of living processes. It is yet another barrier that separates us from its understanding. Goethe has taken the first step, with a new possibility to cross this threshold. He remained entirely in the realm of morphology [external form, or structure]. Now the time has possibly come, that also for the chemical processes in living matter a step further may be made. The experimental work shall thereby go hand in hand with an inner development, such as is given in Anthroposophy. Whenever one omits to do this, one's work can fall into danger. It is not for nothing that Rudolf Steiner has repeatedly said, 'in order to penetrate the realm of the living, first the laboratory work-bench must become an altar'." Thus through what is essentially a meditation based on a living memory of the experimental observations, such as were made by Holleman, an insight may be attained into the quality, or essence, of the idea lying behind the phenomena under study.

Now, in Anthroposophical science consideration is often made of so-called higher processes or forces, observable only by means of our alleged higher senses [our lower senses being those of sight, touch, smell, etc]. As an example, such higher senses are claimed to be used to "read" the blood crystallization plates used in Anthroposophical medicine in the diagnosis of a range of illnesses and conditions. Despite Wim Holleman's anthroposophical background, his own blood crystallization experiments conducted for a short time after a training at Dornach in Switzerland (probably some time around the late 50's or early 60's) did not follow Anthroposophical dogma. Until then (and still to the present day; personal communication from Machteld Huber, March 1996, Bolk Institute) such diagnoses are made by means of both our lower and higher senses, the last of which the majority of us are unaware (i.e. they are something the existence of which I feel unable either to confirm or deny). Conventional photography was considered unable to record any "super-sensible" images. Drawings, made by the highly specialised individuals able to "read" these plates, were the only means of recording and communicating their observations. Holleman considered such drawings not to be accurate, or objective enough. Indeed he believed that the blood crystallization patterns could be photographed and that our lower senses alone, in this case sight, could be used in the diagnoses of a range of conditions, including cancer. He was, however, not sent the blood that was needed for such a research programme.

The above anecdote was included here to throw light on the experimental procedure used by Holleman in his transmutation research with Chlorella. Other researchers (e.g. Haushka, 1983, and Kervran, 1972) believed that biological transmutations were best attained by maximising the "vital forces". Thus natural rainwater should be used rather than distilled tap water; the growth medium should be natural in origin rather than inorganic chemicals; lighting should be natural daylight, or direct sunlight rather than artificial, electric lighting; also the phases of the moon were implicated in the chances of a transmutation occurring. Holleman did not even acknowledge the possibility that such factors may have a role to play. His experimental laboratory procedure was (at least outwardly) entirely conventional.

Holleman was a slow [his family called him "Snail" because he was never known to run, but even so, he was always on time], careful, meticulous man, who was his own worst critic. He preferred to work alone (he allegedly had a previous bad experience with the scientific integrity of a fellow scientist on a previous biological transmutation project). Because of this, and his strong faith in the existence of biological transmutation, he appears to have become almost fanatical about his Chlorella research [one week after retirement, much to the disappointment of his wife, he was back to work on a regular basis, and continued to do so for the following 15 years].

At about the same time as, or shortly after, his first discovery of Steiner's Biodynamic Agriculture course, in 1933, he also obtained knowledge of and eventually a copy of the work of Herzeele (1876-1883). This large body of research had only just been rediscovered in 1930. Holleman briefly reviews this work in section 4.2. What Holleman didn't record was that he had attempted, on a number of occasions prior to his Chlorella research, to replicate some of the extraordinarily large gains and losses of chemical elements in germinated seeds and root cuttings. Holleman's obvious technical familiarity of both present-day and 19th-century chemical analysis led to his finding fault with all of Herzeele's significant results. In fact every one of the positive claims for transmutations recorded by other researchers that were investigated by Holleman [and also by myself, see section 4] proved to be inconclusive. (It is interesting that such literature reviewed by Kervran was accepted by him without criticism). Nevertheless Holleman wrote, in an undated copy of a letter of some 14-16 years after his first Herzeele replicate experiments (late 1940's?), that despite a whole catalogue of analytical sources of error that he had identified, he felt that Herzeele was clearly a technically competent chemist for his time and that he still felt able to believe his (qualitative rather than quantitative) results. Herzeele, apparently, had been inspired by Goethe, and the quote "Der Boden wachst mit der Pflanze" [The soil develops with the plants] was, according to Wim Holleman's daughter, Sophia (personal communication, February 1995), often quoted by her father.

A detailed, critical review of Holleman's practical work is essential to a full understanding of the results attained and for their future replication.

The last four of the six main experiments that are described here were all essentially attempts to reproduce the apparent decrease, and subsequent return, in the potassium content of a closed culture of Chlorella vulgaris as observed in experiment II.

The Chlorella cultures, in order to help ensure reproducibility of the results and comparability of parallel running cultures, were needed to be kept at a constant temperature. Whilst steps were taken to minimise temperature variations within the culture chambers, no comparative experiments were conducted to examine temperature effects on growth.

The agitation by the shaking bath in experiment II was also variable; either insufficient or, due to resonance, waves were created which resulted on one occasion in the loss of culture fluid from the dishes. It is also possible that water from the water bath entered the culture dishes.

The culture vessels, both dishes and tubes, were of quartz glass. Under certain conditions (heated steam), corrosion of the surface of the quartz glass took place [see sections 6.3.2 and 9.1.4.2]. The resulting silicate suspension may have adsorbed potassium from solution causing a decrease in the measured amount of this element. An experiment to investigate such possible adsorption was never completed.

All measuring glassware was tested and calibrated for accuracy and temperature variations.

All chemicals were of "analysis quality". A published list of trace elements present in the "suprapure" nitric acid was quoted by Holleman. This showed copper concentrations to be at potentially toxic concentrations (over the complete course of an experiment). No steps appear to have been taken to either deal with or investigate this further.

The supply of carbon dioxide gas to the cultures was highly variable. The effects of this on growth and health of the Chlorella cultures was not examined. In the latter experiments where the gas stream was used for agitation no comparisons were made between the efficiency of this method and of the shaking bath. Sinking and clumping was a continual problem preventing the formation of a completely homogeneous culture. Foaming proved to be another problem with losses of algae due to their deposition on the sides of the tubes. The use of Vaseline smeared thinly on the gas inlet tubes largely solved this problem. [See sections 6.3.2 and 8.1.2.5].

From the literature obtained by Holleman on this subject it is clear that not only the light intensity but also the type of light source has a profound effect, not only on the growth, but also on the physiology of a Chlorella culture (e.g. Tipnis and Pratt, 1960). It is not clear whether the motive for the switch from a tungsten filament bulb to fluorescent tubes was purely one of convenience. The light intensity, for experiments V and VI, was however considered in some detail on the basis of the literature on the subject. The light intensity received by each culture tube was found to be equal. See also section 9.1.2.1.

A number of measures were implemented, including a comparison of the effects of autoclaving as opposed to ultra-filtration of the nutrient solution. No significant difference between the two methods was found, based on algal growth. See also section 8.1.2.7.

The nutrient solution used throughout was that of a slightly modified solution of Kuhl [section 7.1.3]. The main modification was that the magnesium content was reduced by 90%. This was done against the initial recommendations of section 3.1.5.1. Nevertheless the idea that a transmutation is most likely to occur in the presence of a deficit of an essential element was for Holleman a strong one. The element magnesium is essential for photosynthesis and thus for the whole "organic process" within the plant.

The production of Chlorella is detailed in sections 7.1.4.1 and 8.1.4.1. The cultures were, whilst often variable in results, highly resilient to a wide range of storage conditions.

Strictly speaking, the methods given in sections 7.1.4.2 and 8.1.4.2 were indirect and direct methods of measuring chlorophyll content. Only with experiment VI was cell number directly counted.

The use of these and dry weight production may all be used to assess the general health and vitality of the algal cultures (see section 10.2.2 for details). Cell size and the number of autospores produced per cell division were not however measured. In one or two experiments associated with Holleman's synchronisation studies, a growth constant was calculated. This was not described in section 9.1.3 as insufficient details were given in his notes.

Holleman did not explicitly make comparisons in his notes between direct and indirect measures of chlorophyll production and/or cell number (dry weight was also measured in one or two cases), though the data were recorded every few days for many experiments.

It is interesting to note that for experiment V the culture volume was reduced to 10% of its original value, as used in experiment II, presumably making accurate measurements more difficult [this potential difficulty was, however, never mentioned]. Nevertheless, it did enable not only duplicate cultures to develop in parallel, but also their equivalent controls to be duplicated as well, in the same culture chamber since more tubes could fit in.

Section 9.1.4.1 gives details of the use of a regular light/dark cycle which on its own was able to produce synchronised Chlorella cultures. The experimental procedure described in the literature also included the synchronised dilution of the cultures in order to attain a constant, low concentration of cells. Such dilutions increase the homogeneity of the immediate environment of every individual cell. However, it appears that regular dilutions were not conducted by Holleman. Any culture dilutions (i.e. by the addition of a measured sample of the experimental culture solution to a calculated volume of new nutrient solution) would also dilute the culture medium in which they were growing. As a consequence any accumulated transmutation effects would also be diluted. See section 6.3.3 for details.

The ashing procedures described in sections 7.1.5.2, section 8.1.5.2 and 9.1.4.2 were increasingly simplified and improved after much experimentation. The alkalisation of the ash was suspected to be due to a reaction between the acid phosphates and the nitrates, but probably involved the polyphosphates which had been formed by the algae and by the heating as well. This resulted in the removal of the acid forming ions resulting in an alkaline solution, which Holleman neutralised by the addition of extra nitrate ions in the form of nitric acid. The formation of polyphosphates by the algae was dependant on the culture conditions which, for instance, was dependant on the amount of nitric acid present, which in turn depended on the amount of polyphosphates present. Thus it was found that the pH results were highly variable. Whilst the subject of many experiments, unfortunately pH measurements were not made during the running of the main experiments to check up on this. See section 10.2.2 for further details.

The need to add extra nitrate ions, in the form of nitric acid, to the redissolved ashes to produce usable nutrient solution meant that there was a significant difference in the chemical make-up of the reconstituted nutrient solutions from the original Kuhl solution. Apparently the phosphates that were originally present in the Kuhl solution acted as a pH buffer solution [i.e. they were able to maintain the pH of the nutrient solutions despite variations in the chemical environment that would otherwise increase or decrease the acidity of the solution]. Kuhl and Lorentzen (1964) further state that Chlorella is unaffected by high phosphate concentrations. Therefore a constant pH environment may be better maintained by increasing the phosphate concentration of the original Kuhl solution, and could be used instead of the nitric acid additions (this was proposed in section 9.1.4.2). It was also recorded, however, that under certain conditions a great part of the ashed Chlorella may consist of phosphate originating from inorganic, condensed phosphates (polyphosphates) which can be accumulated by Chlorella to a great extent (see Kuhl, 1960 and 1962b for further details). Such polyphosphates were implicated by Holleman as causing great difficulties in chemical analysis (see especially section 10.1.5.4); thus it may be for this reason that Holleman used nitric acid. It was also quoted from the literature that increasing the nitrate concentration increases the culture density. See however, section 9.2.2.

It is worth noting that all changes in the reconstituted nutrient solution would accumulate; thus the volatilisation of any chemicals during ashing could also have an effect on algal health. See section 10.2.2.

A number of different controls are described in sections 7.1.5.3 and 8.1.5.3. The conclusion that I have come to was that Holleman considered it important that the control treatment should be as near to identical as possible as those of the parallel growth cultures, except that the control algae should be dead. Exclusion of light and heat treatment were two successful methods; the use of selective poisons were also considered but not tried. However, in experiment V, whilst the experimental cultures and the controls ran (in general) simultaneously, the particular control and growth pairs that were to run for the same number of cycles did not run together. There was not enough room in the culture chamber for them to run at the same time. This decision lends further weight to the idea that Holleman did not consider external (cyclic) influences such as the phases of the moon [or other so called cosmic influences] to be relevant (see Hauschka 1983). The 6 cycles of experiment V for example, despite the fact that each growth cycle was only 14 days long, lasted a total of 5 months (not including a planned 1 month break and a further 3 months hospitalisation). Furthermore, it took another 4 months for the ash solutions to be analyzed; a grand total of 13 months from start to finish. See section 10 for a further consideration of Holleman's outlook on the subject.

Details are given in sections 6.2, 7.1.5.4, 8.1.5.4, 9.1.4.3 and 10.2.3. The main elements that were analyzed - with varying degrees of success - were potassium, sodium, calcium, magnesium, nitrogen (or nitrate), and phosphorous (or phosphate).

The reason why potassium was chosen as the main element to be investigated was never stated; there are, however, very strong practical and theoretical grounds for its having been chosen. Historically it has often been implicated as taking part in transmutation phenomena. It is also easily analyzed to a high degree of precision by flame photometry and capable of independent analysis by the Kalignost method. The use of sodium as a neutral reference was considered to be a an extremely clever one by Michel Haring (personal communication, November 1995). The agreements between the controls of experiment II (see section 7.2.1, figure 3 and figure 4) of both Kalignost and flame photometry for potassium and also with the sodium gave strong support for the methods chosen by Holleman. It also prevented the easy possibility of analytical error for the experimental results. Nevertheless, I have considered it my duty to attempt to find such an error (see my foreword to this work, as well as sections 10.2.3 and 11). The potential error in potassium content measurements, that may have been the result of the necessary filtration of any precipitate found in the ash solution before analysis, remains unproven. The improved hydrolysis of the ash in subsequent experiments meant that such precipitates were no longer a problem.

The analysis of calcium and magnesium proved extremely difficult. The presence of polyphosphates apparently hindered their being analyzed. Nitrates and phosphates were analyzed with varying degrees of success in an attempt to understand the chemistry of the ashing process. This [presumably] led to an improvement in the hydrolysis of the polyphosphates. I found it difficult to follow the procedures adopted by Holleman and to assess the success of his analyses.

Holleman quotes Steiner (1924) as stating that it is possible for potassium, via an unknown intermediate step, to transmute into nitrogen. Holleman put foreword the suggestion that this intermediate step could be the noble gas, argon. Thus Holleman refers elsewhere to Picket's analysis of argon given off from yeast. The presence of argon in biological material would conventionally be considered as coming from the radioactive decay of the potassium isotope K40.

Iron and sulphur analyses were considered but were never done [sulphate was attempted but unsuccessfully]. Carbohydrate and protein were also considered for analysis, but no records have been found of this being done.

These are given in sections 7.2.1, 8.2.1 and 9.2. The results of experiment II are further considered in sections 10.1.5.4, 10.2.3 and 11.

The anomalous results of the sixth cycle of experiment V (section 8.2.1) are noted here as being associated with two different calibration solution values and were discounted by Holleman.

The associated measurements, also given in section 8.2.1, of the potassium contents of the 5 remaining (at the time of the analysis) inoculation cultures and of the 2 Kuhl solutions, with and without hydrolysis are also worth noting. The higher than normal values of the inoculation cultures may most simply be explained by the fact that they came directly from the Biophysics department. Thus the potassium content of their nutrient solution was presumably slightly higher than that prepared by Holleman. The lower values of the Kuhl solutions were, as expected, approximately 4% lower than that of the potassium value recorded for the cycle one cultures (see section 7.2.1 for further explanation). These results also clearly demonstrate that the hydrolysis procedure, of itself, has no effect on the measured potassium concentration.

The clumping and associated sinking of the algae could only partly be due to insufficient agitation. It may be, apparently, a sign of ill health. The importance of its occurrence with regards to the maintenance of a homogeneous cell population is obvious. Those cells on the inside of a clump do not have as good an access to light and nutrients as those on the outside. It is interesting that the low starting densities of the cultures associated with experiment VI were associated with low levels of clumping. Clumping and sinking also made it at times impossible to obtain a valid optical density reading.

It is perhaps here a good point to consider the general population dynamics of a Chlorella culture. The growth curve, of cell density against time, is a classic "S" shaped curve. The initial portion of the curve shows exponential growth; the middle section, as the individual cells start shading each other, is approximately linear; as growth continues they proceed to use up the available nutrients and so growth slows, eventually, to zero. Holleman found that this whole process took approximately 14 days. Growth would thus be limited, firstly by shading and then by the first (essential) element to be used up. This was intended to be magnesium, in the hopes of forcing the Chlorella to replace it by means of a transmutation. The plotting of such growth curves for cultures grown under different conditions could enable a number of different factors to be quantitatively considered and compared, i.e. growth rate, effects of growth limiting factors, etc.

The dilution procedure that may have been used in experiment VI was a compromise to enable maximum growth and to accumulate any evidence of transmutations having occurred. These were consistently the two objectives that were to be striven for in the experimental design. Measures of "growth" are varied and were only considered by Holleman in very general terms. Holleman normally refers to "growth" when referring to cell population growth. Only with experiment VI is growth also used to refer to the increase in size of an individual algal cell, rather than the growth in size of the total number of cells in the culture. Holleman referred more than once to Steiner's reference to the "organic process" upon which transmutations were supposed to depend. Steiner (1924) refers to the "organic process" in very general terms. The two types of growth mentioned above are not necessarily equally affected by the same culture conditions. In fact, a casual glance at the literature demonstrates that Chlorella presents a wide range of physiological responses to different culture conditions. On the basis of Holleman's early Chlorella results, he considered that transmutations may very likely be associated with particular physiological, or "organic" processes. Thus the synchronisation experiments associated with experiment VI were initiated.

Holleman, a short while earlier, had observed that a heat treated control of experiment V had started to grow and in fact showed remarkably strong growth. [This control culture was started before Holleman's extended period in hospital, during which time the cultures and controls were stored. It was on his return that the beginnings of cell growth were noted. My first reaction was the possibility of contamination which Holleman did not, on paper, consider.] A small experiment (section 6.3.2) proved that insufficient heating would fail to kill all of the algae. After further consideration (taking into account that this aberrant control culture grew exceptionally strongly) Holleman came to the conclusion that perhaps the assimilative growth process was not responsible for a change in the total [Holleman's emphasis] mineral composition. Possibly it was a "dissimilation" process, ie a breakdown or decay process [?] that was responsible for the occurrence of biological transmutations. As support for this idea he referred to Steiner (1924); there he describes compost [Holleman repeatedly uses the inappropriate term "manure"] formation with the aid of the [biodynamic] preparations 1-7. The problem for Holleman was how to follow this compost/decay ["manure"] process in vitro, in an aqueous solution. He considered the warming or poisoning (by chlorine, etc) [presumably] of a Chlorella culture. By means of this hypothesis then, the exceptionally strong growth of the heat treated control may be explained as follows: the heat treatment, which failed to kill all the algae, caused the majority of the algae to decay. This decay process may have been accompanied by a transmutation phenomenon which could have been responsible for increasing the amount of the growth rate limiting element of the nutrient solution; this would enable an increase in Chlorella growth to take place. [It is worth noting, however, that later, "normal" cultures also attained such growth densities.]

This consideration that the assimilation process, which only occurs in the light, may not be responsible for transmutation phenomena helps explain why Holleman considered the dark cycle of a synchronised culture, i.e. the reproductive phase worthy of particular attention. It is to be noted that the consideration that photosynthesis [assimilation] and growth occurs only in the light and that reproduction and respiration [dissimilation?] occurs in the dark is a gross simplification of the enormous variety of biochemical or physiological processes [i.e. the overall "organic process"] that takes place in such a small (4-10 thousandths of a millimetre) single celled organism. I often had the feeling that Holleman's awareness was directed mostly to the chemical phenomena with which he was familiar. Of the almost infinite complexity of the workings within such a tiny cell, Holleman made little comment.

He wrote in a letter, in 1991, that the aim of his research was to enable "a known quantity of mineral substance to be exposed to the influence of an organism, not once but six times". From this I understand the object of his study to have been the "known quantity of mineral substance". I have the impression that most of the large quantity of Chlorella literature was only reviewed by Holleman during his later synchronisation research.

The potential error that may have caused the initial disappearance and subsequent return of potassium (section 7.2.1) was not included in Holleman's privately circulated German language report. Over the subsequent 7 years of experiments, a few addressed the subject of what Holleman referred to in his laboratory notebooks as "trivial reasons" for such a result (see section 6.3.2). None of them were ever completed: other experiments were considered more important; objections were made to other such experiments testing for "trivial reasons" on a variety of technical grounds [some entirely valid, others less so]; and arguments were also found so as to make the remainder appear unlikely to be relevant [again some of this reasoning was not always clear when referred back to the original aims of these tests]. In such cases I like to remember the sentiment expressed by the great fictional detective Sherlock Holmes, that everything must be investigated, no matter how improbable. These tests were not (unfortunately) carried through to completion. He allowed his own personal judgement (conscious or otherwise) to get in the way of a thorough investigation.

The potential error mentioned earlier was that the ash solutions to be used for chemical analysis were sometimes filtered due to the appearance of an insoluble precipitate. It is possible that this filtrate contained a small but significant amount of the potassium. The formation of this precipitate, which was not dissolved by treatment with HClO4 for ash analysis, was infrequent and unpredictable. However, the improvements in culture conditions, ashing and ash hydrolysis procedure successfully reduced the occurrence of such a precipitate. Thus such an [hypothesised] reduction in the amount of potassium available for analysis was very much less likely to occur. So it may have been that the subsequent attempts to repeat these results have proved unsuccessful. [The experiment that unsuccessfully attempted to test for the most probable "trivial" explanation (as described in section 6.3.2) was never repeated; in contrast, the large experiments to replicate the suspected transmutation which was recorded in experiment II were, however, repeated a total of four times.] Therefore the possibility of a "trivial reason" as apposed to a transmutation having occurred must still remain a possibility. [At the time of writing the relevant laboratory notebooks are unavailable for a further investigation of this subject].

The letter quoted from at the end of the last section (10.2.2), dated 2 years after his last recorded Chlorella notes, states the doubts which he had after the unexplained potassium disappearance and subsequent reappearance, of the possibility that a brutal error had been made. This echoes a letter written some 40 years previously expressing the grave doubts that he had had when he, for the first time, failed to replicate the results of Herzeele's transmutation work on a wide variety of plant material. Again, he [eventually] does not allow this to get in the way of his faith in the possibility [my emphasis] of the existence of biological transmutations. The absence of a proof, proved [or disproved] nothing. At the age of 86 he expressed the desire, despite his no longer having access to a laboratory and his increasing infirmity, to finish the work that he had started, including writing up his earlier replications of Herzeele's work. This was, unfortunately, work that now is for others to continue.