The methods of marking from 1781 onwards require a good deal of close examination to determine what nuances of difference were introduced in the engraving of punches to outwit unscrupulous silversmiths.
As I have stated elsewhere it becomes obvious that marks on teaspoons could be easily "let in" on the foot rims of jugs and other similar items so that it was of some importance that teaspoon marks could be differentiated from other marks of similar size.
Apart from the omission of the leopard's head the main difference between the teaspoon mark and that used on other plate is in the shape of the punch in which the sterling lion is engraved. Whereas on all marks, other than those designed for tea and other small spoons, the lion is in a rectangular box with an ogee base and canted top corners the lion on these smaller spoons between 1781 and 1785 is in a roughly oval outline. (Fig 1)
For plate other than teaspoons more than one punch, and indeed stub for use in a fly press, of the same size was made for each year and these punches and stubs were designed for use on specific items and were, usually, not interchangeable although, oddly, this rule does not hold good for sugar sifter spoons.
The punches used on sugar sifters varied and there is no obvious reason for this variation.
Fig. 2 shows two fiddle pattern sifters of the early 19th century. The one bearing the date letter "F" for 1801 is five and nine tenths inches long and the one bearing the date letter "C" for 1818 is five and seven tenths inches long. They are both therefore virtually the same size and were both marked by means of the fly press but on one the teaspoon stub has been used whilst the other has been marked using the large spoon stub!

Fig. 1: Teaspoon hallmarks for 1784/5 showing the oval outline to the lion. (note the duty mark on the upper spoon which shows that this spoon was assayed after 1st December 1784 whereas the lower spoon was assayed between 29th May and 30th November 1784.)

Fig. 2: Top marking on sugar sifter spoons showing different fly press stubs used to mark spoons of similar size

From 1786 until 1805 the outline to the lion becomes rectangular with a curved base and canted top corners (Fig. 3) although in the years 1792, 1793, 1794 the punch outlines are not well defined. From 1806 onwards the outline adopted is the same as for other punches; i.e. rectangular with an ogee base and canted top corners. (Fig. 4)

Fig. 3: Teaspoon hallmark showing the outline to the lion in use between 1786 and 1805 (NB: This outline can appear almost oval especially in 1792)

The outline of the date letter during this period consistently has a curved base and canted top corners, (Fig. 3) but variations in the overall shape between square and rectangular give rise to an almost egg shaped outline in some years, notably 1792 ( Fig. 5).
From 1805 until 1809 it, like the sterling lion, adopts the standard canted top corners and ogee base appearance (Fig. 4). There is no obvious explanation for the poor execution of the engraving of the outlines to both the lion and the date letter in the years 1792, 1793 and 1794. In neither are the canted corners well defined but neither has been engraved in an obvious ovoid form which would suggest an intentional deviation from the mark outlines of the other years in this sequence. It could be just a matter of pressure of work on the engraver and it has to be said that the margin between canted corners and a curved appearance is somewhat narrow especially when the mark has become worn. Experimentation was continuing and in 1792 the workload falling on the engraver, John Pingo, from the Company had become so great that he had to give up his own business in order to cope with it and this gave rise to a petition in which he enumerates this growing workload. The following is an extract from that petition sent to the Court of Assistants of the Goldsmiths' Company by John Pingo in 1792 which throws some light on this:
" ....59 Marks only were delivered in 1790 - 240 Marks delivered in 1791 - 262 Marks delivered this day besides 39 delivered since January last, making in all 301 Marks delivered in 1792"(see note 1))
Was he perhaps cutting corners both physically and metaphorically?

Fig. 4: Teaspoon hallmark showing the outline to the lion in use from 1806/7 onwards and the outline of the date letter from 1805 until 1809.

Fig 5: Teaspoon hallmark showing the date letter outline apparent in 1792 (note that the lion outline can appear virtually oval in this year)

There are obvious variations in the shape of the outline to the duty mark in specific years but these are concerned with the Government's increase in the amount of duty payable and for information on this I must refer my reader to the excellent work on the subject published by A.B.L. Dove F.S.A. in "Antique Collecting" ( September 1984 ) and reprinted in 'Silver Studies', the journal of The Silver Society, number 22 (2007).

The only other observations to note concerning the duty mark are that from 1784 to 1832 the mark was engraved as the monarch's bust and from that date onwards until duty on plate was abolished in 1890 the mark was engraved as the monarch's head couped at the neck. The incuse mark of George III and the mark of Queen Victoria face to dexter and all other duty marks face to sinister.
There is, however, one important feature to note in connection with the duty mark and this concerns that used during the reign of Geo IV (1820-1830). It appears that John Smith, who was by then the Company's engraver, attempted to represent the 'quaffed' hairstyle of Geo IV but in some years, notably 1830, has executed the engraving so poorly that, if the bust of the monarch is recognisable at all, it almost appears to be crowned. (Fig. 6). It is not known exactly when Smith's eyesight began to deteriorate but it had got so bad by 1839 that he was forced to retire and the position of engraver to The Goldsmiths' Company was taken by William Wyon. It may well be that poor eyesight is the explanation for this misleading engraving and the unwary may be tempted to think that it is a forgery. It is, however, just one more small peculiarity to be aware of.

That in any given year the marks have been impressed in such a way that they must sometimes be read with the bowl of a spoon on the left and sometimes with the bowl on the right is, I believe, of no significance since, from the point of view of transposition, once the marks have been cut out of the spoon it makes no difference which way round they were stamped. The order in which they relate to each other, however, may be of significance. The date letter on small spoons precedes the lion between 1781 and 1785 and follows it from then on. The duty mark is always the last in the sequence on tea spoons whereas on sugar tongs it is the first in the sequence (Fig. 7). On larger spoons the date letter is first in line followed by the lion with the leopard's head next. The duty mark is always last so that during the incuse period when it was applied before the hall marks the latter often had to be squeezed in between it and the maker's mark. From 1786 when the duty mark was incorporated with the other marks on the fly press stub the sequence on large spoons became the same as on other plate, namely; lion. leopard, date, duty (Fig. 8).

Notwithstanding all this endeavour on the part of The Goldsmiths' Company and that by the statute 13 Geo. III cap.59 the penalty for transposing marks, as with other types of fraud, was fourteen years transportation, the practice appears to have continued and it is not surprising, therefore, that in 1805 the committee resorted to marking large flatware in the way in which it had been marked before 1781 with the lion at a right angle to the other marks. The layout of the marks also adopted a vertical format (Fig. 9).

Fig. 6: Hallmark for 1830/1 showing the engraving of the King's bust duty mark with crown like hairstyle.

Fig. 7: Press mark on tea tongs showing the duty mark first in the sequence

Fig. 8: Tablespoon by George Wintle London 1801/2 showing the order of marks from 1786/7

It must have been considered that the omission of the leopard's head on teaspoons and sugar tongs would suffice to protect against their use for transposition but alas this was a forlorn hope. Helmet cream jugs mounted on square plinths can be found marked along the foot rim. Such marks are not only in the wrong place but will be seen to be missing the Leopard's head mark indicating that that part of the foot rim started life as the stem of a teaspoon and that the jug was never assayed.
In 1810 the layout of the marks on teaspoons and sugar tongs became the same as for larger flatware, i.e. a vertical stub was used but with the marks in the order lion, date, duty and although the lion outline retained its ogee base that of the date letter reverted to the curved base appearance. It was not until 1821 that the decision was taken to add the leopard's head to this sequence which then became leopard, lion, date, duty.

There are two other features affecting hallmarks which took place during the first half of the 19th. century and it is difficult to see how either could have been connected with security although just what they were connected with I have been unable to determine.
Firstly, in 1821 the Company's engraver, John Smith, was instructed to make "experimental" stubs as well as the regular ones. The alterations chosen were to make the sterling lion passant instead of passant guardant and to deprive the leopard, which appeared for the first time on tongs and teaspoons in that year, of its crown.
This arrangement applied only to the vertical marks used on flatware. When the new marking year started the new regular stubs were used but the committee soon approved the experimental press marks and they therefore came into permanent use. Thus 1821 is known to collectors as the year in which the leopard lost its crown but because the introduction of the new stubs took effect after the marking year had started there are pieces stamped in that year with either the crowned or the uncrowned leopard. (Fig 10)
In 1822 all stubs and punches were engraved in this new way and this format has remained in use for marking all plate since then.

There is one other peculiarity which was introduced on the experimental stubs and that is that the outline to the lion became a rectangle with canted top corners and a curved base on large flatware, Fig. 11, but the ogee base to the lion outline remained in force on small spoons and tongs until the end of the 1821/2 marking year. The new, curved base, outline came into use on these items in the marking tear 1822/3 and remained in use on these until1826. The lion outline reverted to the canted top corners and ogee base format on large flatware in 1823 and 1824 but returned again to the curved base in 1825. From 1826 onwards it had the ogee base which became the standard pattern for all stubs and punches.

The other, somewhat baffling, feature is what has been described as "the punk leopard". Between 1834 and 1839 the leopard has 'hair' which appears to stand on end in what can only be described as a 'punk' style (Fig 12). It has been suggested that this feature was an accident accounted for by Smith's failing eyesight but the hairs are much too carefully and neatly engraved to be an accident and besides they are reproduced in exactly the same way in each of the years mentioned.

Fig. 9: Large spoon of 1805 showing vertical marking

Fig. 10: Hallmarks on sugar tongs showing both the crowned and the uncrowned leopard's head used in the 1821/2 marking year.

Fig. 11: Tablespoon possibly by Edward Farrell London 1821/2 showing the marks used on large flatware (uncrowned leopard, the lion passant and the changed shape of the outline to the lion).   Note the curved base appears to have been created by canting the bottom corners.

Furthermore it seems that Smith was experimenting with this form of the leopard from the moment it lost its crown in 1822, long before his eyesight is known to have been failing. During the later 1820s he engraved the leopard on the punches used on large plate in two forms. In one the leopard is bald and clean shaven and in the other it has whiskers and the suggestion of hair. It is likely that pieces will be found marked in either way but both are genuine. This peculiarity does not appear on the vertical stubs used on flatware. I thought at first that this was a satire by Smith on William IV who was known to have a 'bouffed' hairstyle and, in fact, was known as Pineapple Head. However he came to the throne in 1831 and was gone by 1837 when Queen Victoria acceded to the throne so that theory hardly holds water. I have puzzled over this for some years and have found nothing in the minute books at Goldsmiths' Hall to throw light on this oddity. I have to admit that I cannot come up with a satisfactory explanation.

Although the early representations of the leopard are quite obviously of a male lion, during the latter part of the 18th century the leopard takes on an almost human appearance and in the early 19th its crown has the three points of a jester's hat making it look like a jester (Fig. 10). After 1821 the leopard's visage takes on a troll-like expression and the most likely explanation for these variations is that the engravers had never seen a leopard! The more modern representation is quite catlike and easily recognisable.

Fig. 12: Hallmarks for 1835/6 showing the 'Punk' hairstyle of the leopard's head