Article 3. Written April 2025. Copyright Gavin Bottrell.
Contact me using info@timewarpgolf.com
How heavy were Feathery Golf Balls?
Shortly after purchasing an old feather ball in 2007 I became aware balls in excellent condition often have numbers written on them, however, mine was in poor condition with no legible markings. In some texts the numbers are referred to as the size, whilst in others they are referred to as the weight. These are two distinctly different properties hence both cannot technically be correct. I took my ball to my local post office and weighed it on their scale – it weighed 45 grammes. The current rules state that a ball can weigh no more than 1.62 ounces, which equates to 45.93g so my feathery ball felt very similar in weight to a modern ball when held in the hand. I’d seen feather balls marked 27 and 28, so clearly these numbers are not the weight in grammes. I started to investigate further and consulted as many texts as I could find that referenced the numbers written on feather balls. Some examples (the bolding is mine):
Kevin McGimpsey1 in 2003 wrote “Golf historians are still unsure which weight system, if any, these early ball makers used”. He references an article written in Golf2 magazine in 1896 “The old ball was stuffed with feathers, and feathers were then, and probably are now, weighed by Troy weight: twenty-four grains to the pennyweight, and twenty pennyweights to the ounce; so that the balls have always been weighed by the same measure as feathers.”
Henderson & Stirk3 in 1978 state “The feathery balls were numbered in size from 25-33 dram weight with the most popular being 26-28. The largest ball known to have been made is a 33 by T.Alexander of Musselburgh and this can be seen in the Honourable Company of Edinburgh Golfers’ Collection at Muirfield. A tiny feathery ball together with a particularly large one just under 2 in. in diameter is also in the Burgess Company’s collection at Barton, Edinburgh.” As shown later size 34 balls exist but given the times Henderson & Stirk’s book is remarkable for the range of information it covers.
The earliest dated reference to ball weight I have found is Grierson4 in 1807 who writes “Balls are used of about 1 1/4th of an inch in diameter, and weighing from 26 to 30 drachms avoirdupois.”
In 1687 Thomas Kincaid5 wrote “Your ball must be of middle size neither too big or too little, and then the heavier it is in respect of its bigness it is still the better.” Whilst Kincaid doesn’t specifically refer to the numbers written on balls he clearly knew that a relatively heavier and smaller ball offers advantages.
I started to wonder how I could definitively work out what these numbers mean, but first I had to learn about the units of weight.
A Very Brief History of Weights.
Over the last one thousand years in the British Isles there have been numerous changes to how weights and measures have been defined. Measures often developed in relation to features of the human body, for example the length of a man’s foot or walking stride. Weights tended to develop around harvested goods and the mass of a simple grain of barley became a basic unit. However, by the start of the 19th Century things had become very complicated. The Scots Magazine6 of 1824 provides conversion calculations between the more common systems in use, namely the Scots Long Measure, the new Standard Long Measure, the Troy and Avoirdupois scales, and the Apothecaries and Avoirdupois systems. To make matters even more confusing Glasgow used a different weight system to Edinburgh, which was also different to those used in other Scottish provincial towns! Over the course of the 19th Century eleven major Acts of Parliament concerning Weights & Measures were passed. Frankly, it is baffling.
From the texts cited it appears the writers believed the numbers on balls were either drachms or pennyweights. However, a dram (sometimes written as drachm) contains a different amount of grains within the apothecary and avoirdupois systems. Pennyweights have a very long history and were named after coins first issued by King Offa in the late 8th Century. Offa’s coins were based on a Roman coin called a denarius, hence one penny was written as 1d, and it weighed twenty-four grains. Thankfully, we can use grains to convert between pennyweights, grains, grammes and drams. Since 1959 (yes it changed again!):
1 dwt (Troy) = 24 grains (gr) = 1.55517384 grammes (g) = 0.877 dram avoirdupois (dr. avdp.) = 0.4 dram apothecary (dr. apoth.)
The Weights of Balls Inspected
In August 2024 I was able to construct the following table for numbered balls I had personally inspected. I reasoned this would easily tell me which weight unit was being used. By this stage I had also measured the dimensions of the balls and was reasonably confident the numbers referred to a measure of weight and not physical size.
Table 1.
The results indicate a correlation between the numbers marked on the balls and the calculated pennyweights but they are not exact. We can see that the Dram Avoirdupois and Dram Apothecary weights do not correlate. I felt I could not draw an absolute firm conclusion without a greater sample of balls, and in December 2024 thanks to The Wilson Family Research Fund and the R&A World Golf Museum in St.Andrews I was able to weigh and record data from a larger number of balls. After returning home I constructed the following table.
Table 2.
Those marked in green show a good correlation between the pennyweights and the marked numbers, those in amber a reasonable correlation, whilst those in red a poor correlation. These results were definitely not what I was expecting to see and my heart sank thinking something had gone wrong! I wondered if I’d taken enough care when weighing the second group of balls – if the scale is not placed on a hard surface then spurious weights can be indicated. I started looking back through the photographs I had taken – nothing looked amiss with my experimental methods – the scale, which I had specially purchased, was on a hard tabletop with nothing underneath. I noticed that all the balls that didn’t correlate at all were weighing much more than expected. I consulted my notes and found that against the heaviest ball of all, made by D.Gressick (ref MLC 04 026), I had written that the stitches appeared “very black” and the seams appeared “gummed up”. (see Figure 1). For the next heaviest, a ball by John Gourlay (ref MLC 04 027), I had written that a sticky residue was present in patches on its surface. Whilst I had not written anything against the ball by Tom Morris (ref MLC 04 029), an inspection of the photograph clearly shows blobs of orange residue. The ball by Allan Roberston (ref MLC 04 030) is darker overall than would normally be expected, and there are even darker areas in the seams.
Figure 1. Ultra-heavy ball by D. Gressick (MLC 04 026) and unusually heavy ball by T.Morris (MLC 04 029). Arrows indicating sticky residue.
Figure 2. Over-heavy balls by A.Robertson (MLC 04 030) and J.Gourlay (MLC 04 027). Arrow indicates patch of sticky residue.
Drawing on my own experiences of making replica feather balls I believe I’ve found the reason why these particular balls weigh more than they should – they have been exposed to linseed oil. (Whilst this has not been confirmed by chemical analysis, I am confident in this assertion. There is a small chance that a spirit-based varnish could also have a similar effect if very liberally applied). The Gressick ball has the number 32 written on it, and so I would expect it to weigh about 53g, but it actually weighs 72g which is a huge increase. From practical experimentation I know that a single coat of oil paint can add as much as 3g, whereas if a ball is completely dunked in linseed oil for a few days its final dry weight can increase by as much as 25g as lots of oil will be drawn through the seams into the feather mass by capillary action. In 1999 I had read in McGimpsey & Neech’s book7 “Balls could be regularly oiled to increase their water resistance, not necessarily by their owners but more probably by the caddies. Well-oiled feather balls even floated and were retrievable from water hazards.” I began to wonder whether these balls are evidence that balls were indeed oiled. From practical experimentation I know know that any ball weighing more than 55g does not perform well so I think the Gressick ball either fell into a pot of linseed oil by accident, or was deliberately dunked in it perhaps as an experiment. It is much too heavy to be used effectively, and this may be why it survived intact as Gressick balls are rare.
The other over-heavy balls by Gourlay, Morris and Robertson weigh approximately 6g more than expected given their marked-on numbers. This magnitude of weight increase can be achieved by two or three liberally painted-on coatings of paint or linseed oil. Notably, all these balls have an MLC prefix to their reference number, indicating they are part of the “Museum Loan Collection” and their past provenance is not clear. If deliberately applying oil over the paint resulted in a performance or durability benefit, then I have no doubt that all ballmakers would’ve done it as a matter of course, but none of the other balls appear to have been coated. Moreover, the paint of the period was linseed oil mixed with ground up white lead, and so more coats of paint will have the same affect and hence applying linseed oil over paint makes no sense. Regarding the point “well-oiled feather balls even floated”, through experimentation I know that all feather balls have enough built-in buoyancy to float, so this is not an added benefit of applying linseed oil. I then researched whether any pre-1900 texts refer to the oiling of balls, as opposed to painting or re-painting, and could not find any mention. As is said in legal circles “the absence of evidence is not the evidence of absence” and so it is not possible to say that oiling of balls never took place. However, it is my firm belief that the oiling of balls did not take place for performance reasons and that it is likely a myth that has gained traction. The re-painting of balls definitely did take place and will be discussed later on.
At this time I recalled my first visit to Royal Blackheath GC in 2008 as I’d taken photographs of two very orange coloured feather balls in their collection. From photographs I could see that the leather grips and wooden shafts of the ancient clubs in their collection also had a distinct orange hue. My conclusion is that these two feather balls, and many of their clubs, have been liberally covered in linseed oil at some time in the past for preservation reasons - something no responsible museum curator or amateur collector would do now.
Figure 3. Feather balls as displayed at Royal Blackheath Golf Club in 2008.
The stickiness I felt when inspecting the overly-heavy Tom Morris ball is very puzzling. Linseed oil dries slowly by reacting with oxygen and can take up to two months to fully cure, but this ball is over 170 years old! It is likely the linseed oil used was of the raw rather than boiled variety as it is notorious for taking a very long time to dry and some have suggested it never fully cures – this appears to be evidence of that.
Considering the seven balls highlighted in amber in Table 2, four of them have calculated pennyweights higher than the marked-on number, whilst three have less. The range of error varies from 0.7 too heavy to 1.4 too light. Of these seven balls two were made by Tom Morris, three by the Gourlay brothers, and two by Allan Robertson. Each maker has rated balls both too heavy and too light so there is no consistent bias relevant to any particular maker which might have been expected if the makers scales or weights were inaccurate. From personal experience I know that a moisture-filled ball can take weeks to fully dry before painting, and several months after painting. If the results had only shown balls being slightly underweight then this would indicate they had been weighed before fully drying out. But why some are over-weight, albeit not my much, is intriguing. I wondered whether errors might have occurred when the balls were being weighed. If you’ve ever witnessed a market trader selling fruit they do not take great care to ensure the balance beam is level, but feathery golf balls are much more valuable than grapes! Following the Golf magazine article in 1896 a correspondent8 wrote “allow me to express a doubt that the feather-ball makers of other days ever used any standard measure in marking the weight of their balls. William Gourlay…used pellets of leaden shot to mark the weight of his balls”. Lead shot comes in lots of different sizes so it may be that Gourlay did use round shot instead of the more common flat brass weights, but I have no doubt they would’ve correlated to pennyweights. The ballmakers, unlike some traders, were not legally obliged to ensure their products met the weight written on their goods; it was a guide for the purchaser. If a ball marked 29 was actually a 26 then an experienced player would notice but I doubt a one pennyweight difference would’ve been. Historical leather artifacts are notorious for causing problems to museum conservators as they are very susceptible to the environment around them. It might be that the small positive and negative variations of up to one pennyweight between actual and indicated weights are simply due to either moisture being lost or gained.
I am in no doubt that the ballmakers would’ve understood they were pennyweights even if their customers didn’t. From commercial reports9 we know that thousands of balls made in St.Andrews each year were sold for use in other towns across Scotland and much further afield. The will of David Robertson10, Allan’s father, lists customers living in America, India and more generally “abroad”. Johnston & Johnston11 highlight John Ross writing home from Philadelphia in 1763 saying “I must desire on receipt you may get bought twenty dozen golf balls of the best St.Andrew’s kind for me from 28 to 32 and 34 drops each.” The authors suggest he actually means drams but as I have shown there is no correlation to this unit. I feel it possible that golfers may have used the term “drop” or even “dram” colloquially and neither knew, nor cared, what the actual unit was. Incidentally, a drop was an official measure of liquid volume, but as it equalled 1/60th of a teaspoon it is clearly not applicable. In December 2024 a batch of five balls were sold at auction in Edinburgh. These balls all carried the owners stamp Dundas of Dundas. What is significant and interesting is that three were made by J.Ramsay of Musselburgh, one by John Gourlay and one by Allan Robertson. This batch of balls provides clear evidence that whilst some golfers likely purchased balls from one favourite, and perhaps local-to-them maker, others were happy to buy from different makers. This reinforces my view that all ballmakers worked to a standard weight measure even if their customers didn’t fully appreciate what unit was being used. Given the relatively high price of balls no maker would want to gain a reputation for producing balls of radically different weight to that expected by a customer. It has crossed my mind whether ballmakers were in the habit of estimating the weight without actually checking them on the scales but from personal experience I know this is extremely hard to do and given the time taken to make them omitting the final weighing step would be ludicrous.
Grierson’s 1807 text is the earliest I have found that specifically mentions the weight of balls and scale used: “twenty six to thirty drams averdupois (sic)” but much of the wording is suspiciously similar to those used by Guthrie12 in 1770 to describe golf. The numbers correlate with what we know, however, as I have shown above the units do not. The number of grains to a pound was set at 6992 back in the 14th Century and was revised to 7000 grains in the 16th Century; not enough to account for being over ten percent out. Trying to understand all the weight and measure changes that have occurred is extremely difficult and requires in-depth expert knowledge. It might be that at some earlier time in history drams avoirdupois were used to classify the weight of golf balls, and that Grierson was somehow aware of this, or he might just have made an incorrect assumption. Having viewed the entries about golf in a considerable number of Victorian encyclopaedias it is evident that plagiarism was rife. However, it is very common to find that some individual words have been changed, seemingly at random, and I think this was almost certainly done to circumvent copyright laws. In 1855 Walsh13 repeats Grierson’s words exactly, but in the 2nd edition of 1865 a typographical error appears; the “26 to 30” changes to “26 and 80 drams”. An 80 ball would weigh three times what it should – imagine hitting that! Another example of why it is important to question and verify the entries in old books; just because it’s old doesn’t mean it’s correct!
My investigations show the numbers on feathery balls are definitively pennyweights. The oiling of balls for performance reasons, in my opinion, is a myth and in my next article I will provide further evidence to support this conclusion. With more information available to us it is sometimes possible to determine where clear errors exist in historical texts. These serve to remind us that the validity of recorded personal memories of long-ago practices, or the provenance of historical golf items for that matter, must be questioned.
In the next part of this series of articles I will answer the question “Why were differently weighted feather balls produced at all as surely an optimum exists?” and I will also provide an example of where a conclusion made in a Victorian golf magazine regarding the making of feather balls is demonstrably wrong.
Gavin Bottrell April 2025
References
1. The Story of The Golf Ball, McGimpsey, K. Philip Wilson Publishers, 2003
2. Golf, A weekly record of “Ye Royal and Ancient” Game. Article 20th March 1896 “The Weight of The Golf Ball II”. London.
3. Golf In The Making, Henderson & Stirk. 1978. Pp44.
4. Delineations of St.Andrews, Grierson. 1807
5. The Diary of Thomas Kincaid. 1687. National Library of Scotland reference: Adv.MS.32.7.7
6. The Scots Magazine of 1824 (pp. 452-456)
7. Golf Implements & Memorabilia, McGimpsey, K. & Neech, D. 1999.
8. Golf, A weekly record of “Ye Royal and Ancient” Game. 28th March 1896 . Letter entitled “To the Editor of GOLF”. London.
9. New Statistical Account of Scotland 1838.
10. Will of David Robertson. Wills and Testaments Ref SC20/50/10 Cupar Sherriff Court
11. The Chronicles of Golf, Johnston and Johnston, 1993, pp 248
12. A New Geographical, Historical and Commercial Grammar. William Guthrie of Brechin. London. 1770.