Threesology Research Journal
Periodic Table Triads

(The Study of Threes)

What if the periodic table starts and ends with triads?

Eric Scerri
Department of Chemistry,
Los Angeles, CA 90095


The purpose of this paper is to propose a new design for the presentation of the periodic system of the elements. It is a system that highlights the fundamental importance of elements as basic substances rather than elements as simple substances. Furthermore the fundamental nature of atomic number triads of elements is put to use in obtaining a new perfect triad by relocating hydrogen among the halogens to give the triad H, F, Cl. An unexpected regularity in the period lengths of successive rows is obtained on rearranging the table to start with the halogens on the left-hand side. The relative virtues of this table, as compared with the medium-long form and the left-step table, are discussed.


Sometime ago I made the suggestion that the best form of the periodic table was the left-step form, as first proposed by Charles Janet, in which helium is placed among the alkali earths.i My suggestion was not motivated by any chemical intuition concerning the element helium but rather by a desire for greater regularity in the form of the periodic table. Echoing other authors, I argued that in addition the left-step table reflects the manner in which electrons occupy atomic orbitals more clearly and effectively than the conventionally used medium-long form table.ii The left-step table (fig. 1) allows one to display the n + l rule very prominently again contrary to the medium-long form.

3 H He 1
Li Be 2
2 B C N O F Ne Na Mg 3
Al Si P S Cl Ar K Ca 4
1 Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr Rb Sr 5
Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe Cs Ba 6
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta W Re Os Ir Pt Au Hg TI Pb Bi Po At Rn Fr Ra 7
Ac Th Pa U Pu Am Cm Bk Cf Es Fm Md No Lr Rf Db Sg Bh Hs Mt Ds Rg 8

Fig.1 left-step periodic table

H.O.B. note: While the author's content is focused on Triads in the Periodic Table, it is not explicitly directed towards the "Threes Phenomena" ideology. As such, in the context of a specific "threes discussion" orientation, let me point out that the above left-step table has three steps.

My justification for ignoring the apparent clash with chemical intuition regarding helium was essentially a philosophical one, which formed perhaps the main motivation in that proposal. Although it is not well known, Mendeleev repeatedly argued that the periodic system is not primarily a classification of the elements regarded as simple substances (Lavoisier’s elements) but, more correctly, a classification of elements regarded as basic substances. I suggested that concentrating on elements as basic substances meant that one could ignore the apparently absurdity of placing helium among the alkali earths since elements as basic substances do not possess properties in the common macroscopic sense. Strictly speaking an element as a basic substance possesses no properties whatsoever but as Mendeleev suggested they are attributed just one property — atomic weight, or in present day terms — atomic number.iii

However, one aspect troubled me in this otherwise elegant scheme. This was a conflict with some work that I have been doing in connection with atomic number triads. Although triads were highly instrumental in the discovery of the periodic system, the concept of atomic weight triads was refuted following the accurate determination of atomic weights.

But as I have argued in a recent book, once one accepts that the more correct ordering principle for the elements is atomic number the concept of triads makes a triumphant return, at least in about half of all conceivable triads in the modern table.iv Using the atomic numbers of chlorine, bromine and iodine for example the middle element is not just the approximate mean of the atomic numbers of the flanking elements but the exact mean.

If one looks for an atomic number triads among the elements helium, beryllium and magnesium within the left-step table one encounters a serious discrepancy. Moreover, the conventional placing of helium among the noble gases gives a perfect atomic number triad. So why would one want to lose an atomic number triad by adopting the left-step table? This was one of the nagging problem, at least to my mind, against the repositioning of helium in the way that is carried out in the left-step table.

A New Approach

I now offer a fresh approach that I believe puts a correct emphasis on the fundamental nature of triads in thinking about and organizing the periodic table.

First a few historical remarks about triads. Perhaps the earliest hints of any numerical regularity among the atomic weights of the elements was discovered as early as 1817 by Dobereiner. He was the first to notice the existence of various groups of three elements, subsequently called triads, which showed chemical similarities and which displayed an important numerical relationship, namely that the equivalent weight, or atomic weight, of the middle element is the approximate mean of the values of the two flanking elements in the triad.

In 1817 he found that if certain elements were combined with oxygen in binary compounds, a numerical relationship could be discerned among the equivalent weights of these compounds. Thus when oxides of calcium, strontium and barium were considered, the equivalent weight of strontium oxide was approximately the mean of those of calcium oxide and barium oxide.v The three elements in question, strontium, calcium and barium were said to form a

SrO = (CaO + BaO) / 2 = 107 = (59 + 155) / 2

Though Döbereiner was working with weights that had been deduced with the relatively crude experimental methods of the time, his values compare rather well with current values for the triad:vii

104.75 = (56 + 153.5) / 2

Döbereiner’s observation had little impact on the chemical world at first but later became very influential. He is now regarded as one of the earliest pioneers of the development of the periodic system. Very little happened regarding triads until twelve years later, in 1829, when Döbereiner added three new triads. The first involved the element bromine, which had been isolated in the previous year. He compared bromine to chlorine and iodine, using the atomic weights obtained earlier by Berzelius:

Br = (Cl + I) / 2 = (35.470 + 126.470) / 2 = 80.470viii

The mean value for this triad is reasonably close to Berzelius’ value for bromine of 78.383. Döbereiner also obtained a triad involving some alkali metals, sodium, lithium, and potassium, which were known to share many chemical properties:

Na = (Li + K) / 2 = (15.25 + 78.39) / 2 = 46.82 ix

In addition he produced a fourth triad:

Se = (S + Te) / 2 = (39.239 + 129.243) / 2 = 80.741x

Once again, the mean of the flanking elements, sulfur (S) and tellurium (Te), compares well with Berzelius’ value of 79.5 for selenium (Se).

Döbereiner also required that, in order to be meaningful, his triads should reveal chemical relationships among the elements as well as numerical relationships. On the other hand he refused to group fluorine, a halogen, together with chlorine, bromine and iodine, as he might have done on chemical grounds, because he failed to find a triadic relationship between the atomic weights of fluorine and those of these other halogens. He was also reluctant to take the occurrence of triads among dissimilar elements, such as nitrogen, carbon and oxygen, as being in any sense significant even though they did display the triadic numerical relationship.

Suffice it to say that Döbereiner’s research established the notion of triads as a powerful concept, which several other chemists were soon to take up with much effect. Indeed, Döbereiner’s triads, which would appear on the periodic table grouped in vertical columns, represented the first step in fitting the elements into a system that would account for their chemical properties and would reveal their physical relationships.


Another author who explored triad relationships and perhaps too enamored with their numerological aspects was Lenssen. In 1857 he published an article in which virtually all of the 58 known elements were arranged into a total of twenty triads, with the exception of niobium, which he could not fit into any triad (fig 2). Ten of his triads consisted of non-metals and acid forming metals and the remaining ten of just metals.

Calculated atomic weight   Determined atomic weights

1 (K + Li) / 2 = Na = 23.03 39.11 23.00 6.95
2 >(Ba + Ca)/ 2 = Sr = 44.29 68.59 47.63 20
3 >(Mg + Cd)/2 = Zn = 33.8 12 32.5 55.7
4 (Mn + Co)/2 = Fe = 28.5 27.5 28 29.5
5 (La + Di)/2 = Ce = 48.3 47.3 47 49.6
6 Yt Er Tb         32 ? ?
7 Th norium Al         59.5 ? 13.7
8 (Be + Ur)/2 = Zr = 33.5 7 33.6 60
9 (Cr + Cu)/2 = Ni = 29.3 26.8 29.6 31.7
10 (Ag + Hg)/2 = Pb = 104 108 103.6 100
11 (O + C )/2 = N = 7 8 7 6
12 (Si + Fl)/2 = Bo = 12.2 15 11 9.5
13 (Cl + J )/2 = Br = 40.6 17.7 40 63.5
14 (S + Te)/2 = Se = 40.1 16 39.7 64.2
15 (P + Sb)/2 = As = 38 16 37.5 60
16 (Ta + Ti)/2 = Sn = 58.7 92.3 59 25
17 (W + Mo)/2 = V = 69 92 68.5 46
18 (Pa + Rh)2 = Ru = 52.5 53.2 52.1 51.2
19 (Os + Ir )/2 = Pt = 98.9 99.4 99 98.5
20 (Bi + Au)/2 = Hg = 101.2 104 100* 98.4*

(*In the original version these two atomic weights have been inadvertently interchanged.)

Fig 2. The Twenty Triads of Lenssen. E. Lennssen, Annalen der Chemie und Pharmazie. 103, 121-131, (1857).

Lenssen also suggested further relationships involving groups of triads. Using the 20 triads in the table above he was able to identify a total of seven enneads, or super-triads, in which the mean equivalent weight of each middle triad lies approximately midway between the mean weights of the other triads in a group of three triads (fig 3).

TRIAD Mean Equivalent Weight  
1 23  
3 23 (23 + 44)/2 = 33.5
2 44  
4 28  
6 ?  
5 47  
9 29.5  
8 33.5 (29.5 + 37)/2 = 33.3
7 37  
H 1  
11 7 (1 + 12)/2 = 6.5
12 12  
15 38  
14 40 (38 + 40)/2 = 38
13 40  
18 52.1  
16 61 (99 + 104)/2 = 101.5
17 69  
19 99  
20 101  
10 104  

Fig 3. Lenssen's supertriads.E. Lennssen, Annalen der Chemie und Pharmazie. 103, 121-131, (1857).

An examination of Lenssen’s tables shows that the triad concept was being somewhat forced. For example Lenssen was not averse to identifying triads on a purely numerical basis, even though some of his groups of three elements do not bear any chemical relationship as in the case of carbon, nitrogen and oxygen. It is probably fair to say that much time was wasted by other researchers in trying to uncover triads where they simply did not exist. Some pioneers, including Mendeleev, made it a point to turn their backs on numerical approaches such as Prout’s hypothesis and the search for triads.xi This attitude certainly seems to have paid dividends for Mendeleev in that he made progress where others had failed to do so.

The problem with triads, and also Prout’s hypothesis, is easy to discern in retrospect. It is simply that atomic weight, which both concepts draw upon, is not the most fundamental quantity that can be used to systematize the elements. Atomic weight such as that of lead, as just seen, depend on the particular geological origin of the sample examined. In addition, the measured weight is an average of several isotopes of the particular element. Atomic number, on the other hand, is fundamental and correctly characterizes, as far as presently known, the distinction between one element and the next.

The adoption of atomic number has an intriguing consequence on triads that has seldom been discussed. It is the fact that that approximately 50% of all vertical triads based on atomic number, rather than atomic weight, are absolutely exact! This remarkable result is quite easy to appreciate by referring to the long-form of the modern periodic table (fig 4).

H 3 He
Li Be B C N O F Ne
Na Mg 2 Al Si P S Cl Ar
K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
Rb Sr 1 Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
Cs Ba La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
Fr Ra Ac Th Pa U Pu Am Cm Bk Cf Es Fm Md No Lr Rf Db Sg Bh Hs Mt Ds Rg              

Fig 4. Long-form periodic table.

By considering elements from rows 1, 2, and 3, such as helium, neon and argon one obtains a perfect atomic number triad is obtained.

H.O.B. note: While this table format also contains three steps, the notion of "level" may not have crossed the reader's mind when deducing the three steps in the left-step version. With the placement of Hydrogen (H) and Helium (He) as singularly placed elements at contrasting sides atop the next series of multiple elements, we are confronted with the need to analyze the three steps in a larger "all-embracing" context. In other words, the three-steps are more easily defined in the left-step model. While some readers may find such comments unnecessary because, in their view the three steps are readily apparent; those interested in the threes phenomena need to be aware that inclusiveness can sometimes lead to rationalization... even if it is not being used in the present case.

He 2  
Ne 10 = (2 + 18) / 2
Ar 18

or from rows 3, 4 and 5, for example,

P 15  
As 33 = (15 + 51) /2
Sb 51  

or from rows 5, 6 and 7,

Y 39  
Lu 71 = (39 + 103) / 2
Lr 103  

Alternatively any triads taken from combinations of elements in rows 2,3,4 or 4,5,6, and so on, do not give perfect triads. The reason why this works so perfectly, albeit in only 50% of possible triads, is because the length of each period repeats just once in the long-form periodic table, with the exception of the very first short period. The full sequence is 2, 8, 8, 18, 18, 32, 32, etc. So if one selects any element then there is a 50% chance that the element above and below the selected element, in the same column of the periodic table, will have atomic numbers lying at an equal interval away from the original element. If this is the case then it follows trivially that the second element in the sequence will lie exactly mid-way between the first and third elements. In numerical terms, its atomic number will be the exact mean of the first and third elements, or in other words the atomic number triad will hold perfectly. All one needs to do is to pick a middle element from the first of a repeating pair of periods. Thus half of all the elements are good candidates for where to begin a triad. This phenomenon falls out mathematically from the fact that all periods repeat (except for the first one) and that the elements are characterized by whole number integers. It would appear that the original discoverers had accidentally stumbled upon the fact that some periods of elements repeat twice. What held them back was that these repeat distances vary in length and, of course, the fact that they were operating with the vagaries of atomic weight data. It is somewhat amusing to think that the ancient notion triads of elements, which was initially so productive and later so strongly criticized, should have been shown to be essentially correct, and that the reason for its being essentially correct is now fully understood.

The aim of the present paper is to elevate the role of triads to an even greater extent. Since triads are now expressed in terms of atomic numbers they coincidentally characterize the elements as basic substances or in other words they characterize the true basis for periodic classification.

Mendeleev’s path to mature periodic system

23 39 85 133
7 or 14 24 65 112

16 or 9 15 20 21

Historians differ regarding the precise attribution of the weight of the element depicted as 7 or 14. According to some it is:

Na K Rb Cs
2 Li? Mg Zn Cd

Kedrov, and after him Dimitriev, conclude that the first entry in the second row should be twice the weight of lithium. In any case it is clear that Mendeleev is groping his way towards a horizontal relationship by examining differences in atomic weights and is starting to see hints of almost constant differences in somer cases such as Rb/Zn and Cs/Cd. This endeavor I claim lies in the same spirit as the search for triads. The only difference being that in the case of a triad one seeks two differences between the weights of three elements rather than just two.

Similarly one finds Mendeleev’s first attempt at a periodic system as presented in a hand written table. If one examines the calculations that he is carrying out one finds again an attempt to compute differences between the atomic weights of elements in the columns of his table. For example Mendeleev writes the number 27 in smaller writing below the symbols for potassium (Zn – K = 65 – 39 = 27) and again below rubidium (Cd - Rb = 112 – 85 = 27).

It appears that, in the space of a single day, February 17th 1869, Mendeleev not only began to make horizontal comparisons but also produced the first version of a full periodic table that included most of the known elements. Moreover, Mendeleev’s overall approach consists of looking at atomic weight differences in conformity with the general principle of triads even though he was not specifically identifying triads in the manner of Dobereiner and Lenssen.

Use of triad like concepts to make predictions.

Mendeleev went to some length to distance himself from the use of numerical relationships such as Prout’s relationship and the notion of triads. However, it is quite clear that many of his predictions of the properties of new elements involve the notion of triads. The triads he considered were sometimes vertical, or horizontal or at times the combination of both vertical and horizontal triads.

In the various editions of his textbook, and in the publications dealing specifically with his predictions, Mendeleev repeatedly illustrates his method using the known element selenium as an example. The atomic weight of selenium was known at the time and so could be used to test the reliability of his method. Given the position of selenium and the atomic weights of its four flanking elements:

  S (32)  
As (75) Se? Br( 80)
  Te (127.5)  

The flanking atomic weights can be averaged to yield approximately the correct value for the atomic weight of selenium.xv

(32 + 75 + 80 + 127.5) / 4 = 79

Dissatisfaction among chemists with reduction to quantum mechanics

I turn to an issue in contemporary chemistry. Although chemists have generally been quick to accept explanations for chemical phenomena based on fundamental physics, there has been a reluctance to accept every reductive move offered by physics. For example, some physicists are willing to place helium among the alkaline earths, because it possesses two outer shell electrons. But this suggestion has generally been resisted by chemists. Indeed this has been the main reason why the, otherwise elegant, left-step periodic table has been rejected by the chemical community.

In chemical education circles there has been a growing awareness of the pitfalls of presenting chemistry from a purely reductive perspective which puts electronic structure to the fore. Moreover there is a growing rejection of the reducibility of the special sciences in the philosophy of science and more specifically of the reducibility of chemistry among philosophers of chemistry.xvi

Would it be possible to design a periodic table that unlike the left-step table does not depend exclusively on the reduction of chemical behavior to quantum mechanics in the form of electronic configurations and the filling of orbitals? I believe that it is possible and that the table that I am about to propose does just that. In fact this table will make little appeal to quantum mechanics and electronic configurations but will treat the existence of triads as the fundamental feature of the periodic system.

The new proposal

Finally I turn to the new periodic table, which I claim restores a fundamental role to triads. Rather than relocating helium to the alkaline earths and thereby losing a perfect triad (He, Ne, Ar), I propose to relocate hydrogen into the halogen group, thereby gaining one completely new perfect triad (H, F, Cl).

In chemical terms this proposal is certainly more conservative and more generally plausible to chemists than the relocation of helium, although this is not the reason for suggesting it here. In addition, the relocation of hydrogen is supported in some respects on chemical grounds and has been suggested previously by many authors. But the mere relocation of hydrogen would not represent a significant change to the presentation of the periodic system and my intention is to go further.

Given the usual custom of starting a written page at the top left hand corner it is worth examining how the periodic table would appear if one were to begin with the first column as the group of elements in which hydrogen now finds itself in the new proposed table. This is shown below in figure 5.

H He Li Be B C N O
F Ne Na Mg Al Si P S
Cl Ar P Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se
Br Kr Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te
I Xe Cs Ba Lu Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po
At Rn Fr Ra Lr Rf Db Sg Bh Hs Mt Ds Rg Uub Uut Uuq Uup Uuh
Uus Uuo

La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb
Ac Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No

Figure 5. New proposed periodic table.

I suggest that the habit of ending the periods with a closed shell of electrons is not an essential feature of the presentation of the periodic system. As Philip Stewart has recently reminded us in proposing his spiral periodic system, the elements form a continuous sequence. Similarly, the habit of displaying metals on the left side of the table and non-metals on the right side is just that, a habit or a convention and also not an essential aspect of representing the periodic system. The left-step table for example dispenses with both of these conventions as does my newly proposed table.

An unexpected bonus

It appears that there is at least one rather unexpected bonus feature of the new presentation. It is that the first two periods have the same length as do all subsequent periods. This introduces an added element of regularity which is absent in the commonly encountered medium-long form with its very short period of two elements which unlike all other period lengths does not repeat (2, 8, 8, 18, 18 ...). In the proposed table the sequence is 8, 8, 18, 18 and every period length repeats without fail. It should be noted that the two sequences of eight elements at the start of this arrangement are not the same eight elements which occur in the second and third periods respectively of the conventional medium-long form table.

The lack or repetition of the first very short period of two elements in the medium-long form has been a source of some concern to physicists and group theorists although some have invented devices in order to recapture the sequence of 2, 8, 8, 18, 18... instead of trusting their initial group-theoretical analysis which suggests a greater regularity. Others have achieved greater regularity via the left-step table which of course solves this problem by displaying the elements as sequences of 2, 2, 8, 8, 18, 18... elements.

Summary of features of new proposed form:

  1. Makes the concept of triads fundamental to the presentation of the periodic system.
  2. Makes Z fundamental to the representation of groups as well as the sequence of elements.
  3. Downplays the importance of electronic configurations in the presentation of the periodic system.
  4. Gains a new perfect atomic number triad (H, F, Cl).
  5. Introduces greater regularity in terms of repetition of period lengths. All period lengths now repeat.
  6. Possesses bilateral symmetry provided the rare earths are excluded from the main body of the table. Bilateral symmetry is recovered, admittedly somewhat artificially, by the careful placement of the rare earths as in figure 5.
  7. Correspondence with electronic configuration is not lost. The hydrogen atom has one vacancy in its outer shell as do the halogen atoms.
  8. What is lost is the notion that a period must end with a closed shell. Since the sequence of elements is actually continuous this should not be regarded as a serious drawback in the proposed table.


Of course I am not under the illusion that chemical educators or governing bodies of chemistry will readily accept my proposal. I am suggesting it in order to promote further discussion on the presentation of the periodic system and because I believe that it rests on deep chemical and philosophical principles. As I suggested in my title, I believe that the periodic table which initially arose from the discovery of atomic weight triads can now be further enhanced by recognizing the fundamental importance of atomic number triads and the more fundamental nature of the elements as basic substances rather than as simple substances. Whereas I previously believed that these aims were best served by the left-step table I am now turning my allegiance to the proposed table shown in figure 5.

Notes and References

  1. E.R. Scerri, Presenting the Left-Step Periodic Table, Education in Chemistry, 42, 135-136, 2005; Some Aspects of the Metaphysics of Chemistry and the Nature of the Elements, Hyle, volume 11, 127-145, 2005; Relative Virtues of the Pyramidal and Left-Step Periodic Tables, in The Periodic Table : Into the 21st Century, D. Rouvray, B. King (eds.), Science Research Press, UK, 2004, p. 142-160.

  2. G. Katz, ‘The Periodic Table: An Eight Period Table for the 21st Century’, The Chemical Educator, 6, 324-332, 2001.

  3. F.A. Paneth. (1962). The Epistemological Status of the Chemical Concept of Element, British Journal for the Philosophy of Science, 13, 144-160. Reprinted in Foundations of Chemistry, 5, 113-145, 2003.

  4. E.R. Scerri, The Periodic Table: Its Story and Its Significance, Oxford University Press, New York, 2007.

  5. The reason why Döbereiner chose to begin his work with oxides is not known. These compounds had recently been isolated in England by Davy and might thus have aroused general interest. In addition, working with the oxides would not have required the isolation of the elements and would therefore present an easier experimental option.

  6. It is worth emphasizing that, contrary to the accounts still found in many chemistry textbooks, Döbereiner’s discovery of triads, whose middle member has approximately the mean weight of the two flanking members, did not in fact concern elements but instead their compounds.

  7. These values were recalculated by van Spronsen using a correct atomic weight for oxygen of 16 instead of the value of 7.5 which Döbereiner used.

  8. A printer's error was probably responsible for this small error in the calculated mean which should be 80.97.

  9. Döbereiner was working with incorrect formulas for the oxides of these elements, MO instead of M2O, with the result that his atomic weights appear to be about twice the currently accepted values.

  10. This seems to be another printer's error and more serious this time since the mean should be 84.241.

  11. Prout’s hypothesis is the view that all elements are composites of hydrogen. This was based on the observation that many elements have atomic weights that are approximately whole number multiples of the weight of hydrogen.

  12. At this date Russia was still using the Julian calendar of the Roman Empire. Most other European countries had switched to the the Gregorian or reformed calendar according to which the date would have been March 1st.

  13. I.S. Dimitriev, Scientific Discovery in statu nascendi: The Case of Dimitrii Mendeleev, Historical Studies in the Physical Sciences, 34, part 2, 233-275, 2004.

  14. This is Mendeleev’s first rough version of a periodic table rather than a fragment table. The block of elements has been circled by Mendeleev to produce the appearance of a tombstone. See figure 3 in Dimitriev’s article (p. 237).

  15. However, Mendeleev did not always operate according to this clear procedure, even in the case of some of his most famous predictions. For example, if his method is applied to predicting the atomic weights, atomic volumes, densities and other properties of gallium, germanium and scandium, it produces values that differ significantly from those Mendeleev actually published. Employing Mendeleev's stated method of taking an average of the atomic weights of four flanking elements around gallium, using the atomic weights available at the time, gives a prediction of 70.9. In fact Mendeleev modified this value to ‘about 69’ by means of a more complicated averaging method which he only explained briefly in a single German publication. The accepted value of the atomic weight of gallium at the time of its discovery was 69.35.

  16. E.R. Scerri, L. McIntyre, The Case for Philosophy of Chemistry, Synthese, 111, 213-232, (1997)

  17. L. Sacks, Concerning the Position of Hydrogen in the Periodic Table, Foundations of Chemistry, 8, 31-35, 2006; M. Laing, A Revised Periodic Table: With the Lanthanides Repositioned, Foundations of Chemistry, 7, 203-233, 2005.

  18. P. Stewart, The Spiral Periodic System, Education in Chemistry, 41(6), 156-158, 2004.

  19. V. Ostrovsky, 'What and How Physics Contributes to Understanding the Periodic Law', Foundations of Chemistry, 3, 145-182, 2001.

Your Questions, Comments or Additional Information are welcomed:
Herb O. Buckland