The vegetable and animal kingdoms are divided into various groups. Formerly, men tried to arrange them in linear succession so that there should be an unbroken line from the lowest to the highest members of the vegetable kingdom, thence to the lowest member of the animal, and onwards up to the highest member of the animal kingdom. Such an arrangement as this, however, was found to be unnatural. Instead of the highest members of the vegetable kingdom being connected with the lowest members of the animal kingdom, it is found that the lowest members of each kingdom are closely connected and that the divergence becomes greater as development proceeds towards the highest members in each kingdom. The doctrine of evolution at once rendered this arrangement natural and easily understood.

Starting from one common point of origin in structureless protoplasm, the various organisms became more and more unlike in each successive stage of development, their resemblance being only recognisable at all in their embryonic condition.

Various attempts have been made to arrange inorganic substances in natural orders. One mode of arrangement is according to their atomic weight - as in the following table:-1

1In this and the following Tables the atomic weights have been corrected.

Element

Atomic Weight

Difference

Element

Atomic Weight

Difference

Element

Atomic Weight

Difference

Element

Atomic Weight

Difference

H

1

K

39

3.5

Y

89.8

2.4

Ce

141

2

Li

7

6

Ca

40

1

Zr

90

0.2

Di

145.4

3.6

Gori

9

2

Ti

49.8

9.8

Nb

94

4

Ta

182

36.6

Be

V

51.3

1.5

Mo

95.5

1.5

W

184

2

B

11

2

Cr

52.5

1.2

Rh

104

8.5

Ir

192.7

8.7

C

12

1

Mn

55

2.5

Ru

104.2

0.2

Pt

195

2.3

N

14

2

Fe

56

1

Pd

105.7

1.5

Au

196.5

1.5

O

16

2

Ni

58

2

Ag

108

2.3

Os

198.5

2

Fl

19

3

Co

58.9

0.9

Cd

111.8

3.8

Hg

200

1.5

Na

23

4

Cu

63.4

4.5

Sn

118

6.2

Tl

203.7

3.7

Mg

24

1

Zn

65

1.6

Sb

120

2

Pb

207

4

Al

27

3

As

75

10

I

127

7

Bi

209

3

Si

28

1

Se

78.8

3.8

Te

128

1

Th

233

24

P

31

3

Br

80

1.2

Cs

133

5

U

240

7

S

32

1

Rb

85.3

5.3

Ba

137

4

CI

35.5

3.5

Sr

87.4

2.1

La

139

2

From this it will be seen that the atomic weights of the different elements form a series, the members of which in most cases differ from one another by 1, 2, 3, or 4. There are few exceptions in which the differences are much greater, and these probably represent blanks which may yet be filled up as our knowledge of the elements increases. This mode of classification, however, reminds us of the Linnaean system in plants, and is artificial rather than natural. In it, the elements which are placed close together possess very different properties, whereas those which are separated from each other present considerable resemblances.

Newland's Table

Member of a Group haying Lowest Equivalent

One immediately above the preceding

Difference

H = I

0 = 1

Magnesium .

24

Calcium . .

40

16

1

Oxygen . .

16

Sulphur . .

32

16

1

Lithium . .

7

Sodium . .

23

16

1

Carbon . .

12

Silicon . .

28

16

1

Fluorine . .

19

Chlorine . .

35.5

16.5

1.031

Nitrogen . .

14

Phosphorus .

31

17

1.062

Lowest term of Triad

Highest term of Triad

Lithium . .

7

Potassium . .

39

32

2

Magnesium .

24

Cadmium . .

112

88

5.5

Molybdenum .

96

Tungsten . .

184

88

5.5

Phosphorus .

31

Antimony . .

120

89

5.687

Chlorine . .

35.5

Iodine . .

127

91.5

5.718

Potassium .

39

Caesium . .

141

102

5.875

Sulphur . .

32

Tellurium . .

128

96

6.062

Calcium . .

40

Barium . .

137

97

6.062

The first important attempt at a natural classification of the elements was made by Newlands in 1864.1 He then arranged them in groups, between the members of which there was a close connection in regard to their chemical properties, and a curious relation in regard to their atomic weights These presented differences which were generally multiples of the atomic weight of hydrogen, and generally equal to, or multiples of, that of oxygen.

A curious relationship had also been pointed out by M. Dumas2 between the members of the potassium group, their atomic weights being equal to multiples of those of lithium and potassium added together.

Li + K

= 2Na, or in figures,

7+39

= 46

Li + 2K

= Rb ,,

7+78

= 85

2Li + 3K

= Cs (133) 3 ,,

14 + 117

= 131

Li + 5K

= Tl (203.7) ,,

7 + 195

= 202

3Li + 5K

= 2Ag ,,

21 + 195

= 210

A similar relation was also pointed out by Mr. Newlands between lithium and the calcium group; as follows : -

Li + Ca

= 2Mg (48), or in figures,

7+40

= 47

Li + 2Ca

= Sr ,,

7+80

= 87

2Li + 3Ca

= Ba (137) ,,

14 + 120

= 134

Li + 5Ca

= Pb ,,

7 + 200

= 207

But Mr. Newland's most important table is the following one, in which he has arranged the elements in ten series : -

Triad

Lowest term

Mean

Highest term

I.

Li 7

+ 17 = Mg 24

Zn

65

Cd 111.8

II.

B 11

Au 196

III.

C 12

+ 16 = Si 28

Sn 118

IV.

N 14

+ 17 = P 31

As

75

Sb 120

+ 88 = Bi 210

V.

O 16

+ 16 = S 32

Se

78.8

Te 128

+ 70 = Os 199

VI.

F 19

+ 16.5 = Cl 35.5

Br

80

I 127

VII.

Li 7

+ 16 = Na 23

+ 16 = K 39

Rb

85.3

Cs 133

+ 70 = T1 203

VIII.

Li 7

+ 17 = Mg 24

+ 16 = Ca 40

Sr

87.4

Ba l37

+ 70 = Pb 207

IX.

V 51.3

W 184

X.

Mo 95.5 Pd 105.7

Pt 195

Seven of these series nearly correspond in their first members with those of Mendelejeff, to whom and to Lothar Meyer we owe the complete development of this mode of classification. Mr. Newlands also pointed out that the eighth element starting from a given one, was a kind of repetition of the first, like the eighth note of an octave in music.4

Mendelejeff has not only greatly developed this system of classification, but has afforded convincing proof of its value by not only predicting the existence of an unknown element, but actually describing its physical characters and chemical reactions - a prediction the correctness of which was proved by the discovery of gallium, and by the agreement of its characters and reactions with those which Mendelejeff had foretold.

The various members of the animal kingdom can all be arranged in a few series : Protozoa, Coelenterata, Annuloida, Annulosa, Molluscoida, Mollusca, and Vertebrata. These series all differ more or less from one another, but a certain agreement is observed between their members, and similarly the elements may be arranged in series.

1 Newlands, Chemical News, July 30, 1864.

2 Dumas, quoted by Newlands, op. cit.

3 The newer atomic weights of Cs, Fl, Mg, and Ba do not correspond so exactly as their old ones with the sum of the other elements.

4 Chem. News, Aug. 20, 1864, p. 94.

Mendelejeff points out, that if we take those elements having the lowest atomic weight, and omit hydrogen, between which and lithium there is a great gap, the seven elements, lithium, glucinum, boron, carbon, nitrogen, oxygen, and fluorine, may be regarded as typical elements forming a series representing the lowest members of seven groups. The next seven elements may be arranged in a similar way : Li = 7: G = 9.4: B = 11: C = 12: N = 14: 0 = 16: F = 19:

Na = 23: Mg = 24: Al = 27: Si = 28: P = 31: S = 32: Cl = 35.5.

To each group of seven elements Mendelejeff gives the name of a small period or series. In each series the characters of the elements vary gradually and regularly as their atomic weights increase. This variation is periodical, i.e. varies in the same way in each series, so that the elements which have corresponding places in each series, correspond also to a certain extent in their properties, and form similar compounds. The atomicity is least in the first, and greatest in the last members of each series. Thus the first members of the series form monochlorides, the second dichlorides, the third trichlorides, and so on.

In the accompanying table R represents radical or element, and Ri indicates that the element is monatomic, so that one atom combines with one of Cl to form a monochloride, RC1. Rii indicates that the element is diatomic, and so on.

But a difference is to be observed between the even and the uneven series. Corresponding members of even series, such as the fourth and sixth, agree with each other, and members of uneven series like the fifth and seventh agree. This agreement is greater than between the members of an even series, such as the fourth, and those of an uneven series like the fifth, although the fifth is more closely placed to the fourth than the sixth is. Thus Ca and Sr belonging to the fourth and sixth series have a greater resemblance to each other than they have to Zn or Cd, which belong to the fifth and seventh series, and these metals on the other hand have a greater resemblance to each other than they have to Ca or Sr. The members of even series are less metalloidal or more metallic than those of uneven series, e.g. Mn of the fourth series is less metalloidal than Br of the fifth series. In the even series the metallic or basic character predominates, whilst the corresponding members of the uneven series rather exhibit acid properties. The members of the even series, so far as we know, form no volatile compounds with hydrogen or alcohol radicals, while the corresponding members of the uneven series do form such compounds.