NGC 111

[Demonperformer]

NGC 111 is a "missing" object from the second Leander McCormick list (LMII), #281. The reported position is RA 0h 21m Dec -03º 14' (1890), the description being "very faint, very small, round, a little brighter in the middle, *8.5, p[receding] 36s, n[orth] 2'; RA uncertain.

Now I take you to the image below, which I came across while viewing a list of galaxies on DSS-1. I was struck by the bright star just above the centre on the right of the image.

This star turns out to be magnitude 9.6 (1.1m fainter than that reported for #281) and has an offset position to the galaxy (precessing everything to 1890) of 32.4s preceding, 1'26" north. At this point I was 'struck' even more.

Of the 23 stars referenced in LMII, 22 of them are actually fainter than the reported figure, by an average of about 2m, one by as much as 3.1m, so 1.1m is well within the range of reasonability. There are also instances of star-offsets being out in RA by 4s (LMII #316 = NGC 647, offset reported as 16s, actually 20s). And the declination offset is close enough to 1½' that rounding it up rather than down is well within the realms of reasonability (Leavenworth only recorded the offsets of stars to the nearest second of time [for RA] and arcminute [for Dec]).

Now, the object in the image was also discovered by Leavenworth on LMII and the description he gives is just "very small, very faint, round" - no mention of the star and no mention of it being any brighter towards the middle, which I found suggestive. If indeed these are repeat discoveries of the same object, we would expect the descriptions to be the same, or at least compatible, and in this case they are. But, more than that, the description for #281 actually fits this object more fully that the other description. It is almost as if there were some kind of interruption and/or bad seeing on the night this discovery was made.

But now we come to the nub of the matter: the position is a very long way off - even compared to the notoriously bad positions provided in the LM lists. So can they actually be the same object?

The errors are suspiciously close to 'round figure' errors (which are easier to explain than smaller 'odd' errors). The declination error is fractionally over 20' out (using 1890 figures). Now, you don't have to go very far into the NGC catalogue (all the way to #6!) to find an error in the reported declination that is over twice that (47'). So, clearly, that is within the realms of possibility.

The real problem is the RA error (which Leavenworth described as "uncertain"), which is almost exactly 1½h out. Now let's not pretend that that isn't a big error - it is - and it requires me to at least provide some sort of hypothesis as to how such an error could have been produced.

Firstly, we should note that there are big 'round figure' errors in the NGC, including at least 3 which have an error of 1h, and one of those (NGC 3518) was discovered using this same LM 26-inch telescope.

Secondly, we should note that it is less than 1m away from being exactly 1½h out (1890). And, as the LM lists only record RA to the nearest minute, this is a lot more likely error than if it were (for example) 1h 25m out.

Well, I do have a hypothesis as to how this error could have happened, and am currently awaiting a response to an email I have sent to the astronomy department at the University of Virginia (which owns the LM scope) to see if the assumptions I have had to make are correct or not. [It should be noted that, even if the answer shows that this explanation is not realistic, there could be some other explanation.]

IF the RA circle on the scope is labelled every half-hour, and IF the markings are such that the '2' in the '1/2' is full size (i.e. it is '1/2' rather than '½') THEN could it be possible that someone, in a hurry or while allowing their concentration to lapse, might mistake '2h' with '1/2h', and then take a reading 9m before that?

½h - 9m would give a reading of 0h 21m - the reported position for #281
2h - 9m would give a reading of 1h 51m - very close to the 1890 position for the galaxy in the image [actually 1h 50m 08.2s]

And, if he (or an assistant) had recorded an erroneous position at the time, could it be that only when going over his notes later for publication, he realized this RA did not fit with the area of sky he had been observing on that (and/or on nearby) nights, but had no idea what the true value was - hence the comment 'RA uncertain' [Why would it have been 'uncertain' at the time of the observation/reading?]

I have run this suggestion past several people experienced in the field of identifying troublesome NGC objects. Dr Harold Corwin (of the NGCIC Project) described it as "the first reasonable idea for this object that I've seen" (which was a very nice comment to get). But, with a lack of any supporting sketches or other supporting documentation, it is going to be virtually impossible to prove. Dr Courtney Seligman (http://cseligman.com/text/atlas/ngc1.htm) has said that it "perfectly fits the description" but (due to the large RA error) it is "a very uncertain possibility".

Whatever the truth (which we will almost certainly never know), I maintain that the image is suggestive and, if we only had the image and the description (without the position) we would look at them and say "that fits". In the mean time, as I wait patiently to hear from UVA, any comments are welcome.

And, just in case anyone is wondering, I have deliberately not stated which galaxy is in the image ... I will do so in a subsequent post, but would be interested in people's views before introducing that bias.

[michael.h.f.wilkinson]

Very interesting post. Looking forward to further expert opinion

[Tzitzis]

A great read. You have an analyst's brain mate (if that makes sense in English lol). It's very interesting and I love the way you think on this one. Hope UVA answers quick and we get one step closer to the solving of the mystery.

[DeepSkyBagger]

PM-ed.

[beka]

Interesting post. So my understanding is that the NGC was compiled from different sources among them LMII and it turned out that #281 in the latter was found to be non-existent and labeled NGC 111 - and you set out on a quest to find it?.... And your belief is that it is the galaxy in the DSS image? This is the first time I am hearing of "missing" objects in the NGC and when I checked NGC 111 in Cartes du Cie, it is described as a Plate Defect. If the labeling on the RA circle of the telescope is confirmed to be like in your theory, then I think  it may be a very plausible explanation for the missing object. It wonder if someone can check the telescope labels - pictures on google suggest that it is still operational for outreach and such.

[Demonperformer]

Hi Beka,

There has been a lot of work done identifying objects in the NGC & IC for many years. Take a look at this. I have emailed University of Virginia about the setting circles on the scope, and am still awaiting a response. Ever hopeful.

Thanks.

EDIT: Just had another thought ... it can't be a plate defect, because Leavenworth made his discoveries visually at the eyepiece, not photographically.

[Demonperformer]

I have heard back from UVA, and it seems that this scope does not have a standard (modern) slip-ring RA circle, but a fixed hour-angle circle. This requires the reading to be adjusted by subtracting it from (or adding it to - for readings east of the meridian) the local sidereal time. In the attached photo, the white paint is a "modern" addition. The original markings (which can be clearly seen between 1^h & 2^h)  are 12 equal length marks, each representing 5^m. It can also be seen that the hour numbers run in both directions (so there are two positions labelled 1^h, one on either side of '0').

So, although my original suggestion that someone misread 2^h for 1/2^h is clearly disproved, there are a number of sources for potential errors.
(1) Simply misreading the position off the wheel by miscounting the number of 5^m marks from an hour mark, getting the direction from the hour mark wrong, or getting which hour mark (clockwise or anti-clockwise from 0) wrong.
(2) Arithmetical errors in calculating local sidereal time from local solar time.
(3) Adding to instead of subtracting from (or visa versa, depending on the object's relation to the meridian) the measured hour angle and the calculated sidereal time.

I have run all this past Dr Corwin, and his response is "There still would have had to have been a whopping blunder on someone's part ... there is considerable evidence of other such blunders in these lists".

Thanks.

BTW, the galaxy in the photo in the original post is PGC 7198 (=NGC 758) and the star is UCAC4 435–002390.

[beka]

Nice follow up, I guess like you said we may never reach a definitive conclusion - but there clearly was a "whopping blunder" because we usually do not expect mag 8.5 objects to just disappear - unless it was maybe a comet.  I tried briefly to get some information on how the LMll list was compiled to check if more information on the observations of #281 was available. If it was a single observation the maybe it was a comet. Do you know if other such "missing" objects for which plausible explanations were determined?  

[Demonperformer]

Like most of the LM discoveries, this was a single observation - but the same is true of the Herschel lists. A copy of the published paper can be found here.

My latest thoughts on this are that (1) the HA for an object passing the meridian is '0'; (2) As this is the point when it is highest and therefore faint objects would be more easily visible, this would be 'prime hunting ground' for Leavenworth et al; (3) 45m [RA] on either side of this would be fairly close and so within this 'prime hunting ground'; (4) All that would be required then is for an error to be made in the recording of E/W, resulting in the wrong sign being applied when the conversion of HA to RA was being done, and that would produce an error twice the original reading which is 90m. All-in-all I think this is actually a far more convincing argument than my original suggestion of misreading 2h as 1/2h. It is certainly simpler and there are other cases within NGC where the original discoveries have included directional and sign errors.

Take a look at the NGCIC project pages for many examples of objects that are not where they were expected to be. These are usually identified because of (1) round figure offsets [eg NGC 3158], or reversal of +/-, which can be simply explained; and/or (2) odd offsets, but the surrounding starfield is obvious once you know what the object is [eg NGC 6].

Thanks.

[FenlandPaul]

Just want to say that's remarkable and very interesting analysis! Thank you for taking the time to post your thoughts.

Paul

[Paul73]

Great thread. I missed this one at the time. 

I’d agree with your updated hypothesis (after seeing the photo). Plenty of opportunities for error. I guess that such errors led to the addition of the numbers.

Even then quality checks should have picked up the errors. Nowadays, with a bit of thought, excel makes this really easy.

Paul

[Demonperformer]

Thanks, Paul.

It seems that the main data pages for the NGCIC Project are no longer working, but Harold Corwin's extensive notes on each object can be found here.

[Demonperformer]

Update:
I have just accessed the latest (2018) copy of Dr Wolfgang Steinicke's ongoing spreadsheet of the Revised NGCIC, and note that he has accepted this identity. Dr Seligman, however, remains of the view that this is a "very uncertain possibility".

Thanks