67P/Churyumov-Gerasimenko. A Single Body That’s Been Stretched- Part 47


The two terracotta lines are a translational match. 


This post interrupts the series on Imhotep crust slides. Pages were being posted to avoid the problem of interruption. They will be upgraded to posts when the Imhotep series is finished. But now the menu bar is too full with pages awaiting upgrade so there will now be some posts interspersed between the Imhotep slides.

The line of boulders in Hapi matches the Seth/Babi perimeter. It’s a translational symmetry: two lines that when slid together, overlap nicely, following the same twists and turns. This discovery was going to be posted after Part 40 because it follows on naturally from the Babi slide and would have led us into the site A area for a few posts. However, Imhotep took over and this plan was dropped. Nevertheless a new Cometwatch post on the Rosetta blog has a particularly good photo for illustrating the boulder line-perimeter match so it’s getting a post. The Cometwatch post is here:


In this Cometwatch post, Site A has been officially named Aswan. This blog will try to use it but will have to reiterate that it’s the old Site A on a regular basis because ‘Site A’ has been used hundreds of times in the previous 46 posts and won’t ever get updated. 

Photo 1- this is the same as the header. The line of boulders in Hapi matches to the Seth/Babi perimeter. Therefore the perimeter slid from the boulder line. 

Right hand terracotta- the line of boulders. 

Left hand terracotta- the Seth/Babi perimeter. 

Photo 2- as photo 1 and includes visible slide tracks in yellow.

Yellow- slide tracks.

Fuchsia- a rift along Hapi almost at right angles to the slide tracks. This was mentioned in Parts 38 and 39 along with a promise of a dedicated post, which hasn’t happened yet as of this post. The eventual post will show multiple photos, including the section in shadow. This representation of the match is rather naive because of the actual match being in the shadow but it is there. This rift explains the lateral translation anomaly along the line of the Seth perimeter with respect to the boulder line. The boulder line is continuous with no such rift and the curved slide tracks show the path of the newly freed section as it slid both back with the perimeter and along Hapi with the rift. 

Photo 3- includes a green section

Green- a section that was missing in photos 1 and 2 because it slid further back from the notional perimeter that kisses the dust of Hapi. This is the first indication in this series of photos that multiple delaminations occurred throughout the perimeter slide. In other words, sections of crust delaminated and slid further on while the section they delaminated from stayed put at the location of the delamination. This leads to multiple perimeter lines that are translationally matched. In this case, only two delaminated layers will be shown, giving two translationally matched perimeters. However, along with the line of boulders, it’s three translationally matched lines (see photo 4).

Photo 4- this includes the cove on the head that matches to the body (Parts 3, 18, 35-37). 

Larger yellow- the cove on the head and its seating point on the body. The last two dots at the top in both cases, sit either side of the ‘napkin’ described in Part 5. The napkin is at one tip of the so-called gull wings. The cove line on the head is tricky to depict here but is fairly accurate. It’s tricky for three reasons: 

(1) Head tipping causing foreshortening.

(2) shadow.

(3) the line runs under the visible rim here at both ends. That’s why the dots don’t follow the rim exactly- we are supposedly looking through the head lobe to its underside where the seating is. Photos 6 and 7 show the underside view. 

Dark green-the rest of the visible head rim and its seating on the body (Parts 1, 5, 7, 16).

You can now start to see that the bottom section of the cove in this view is a translational match to the terracotta perimeter and to the boulder line. Photo 5 elaborates on this. 

Photo 5- isolating the translational matches

The cove, the perimeter and the boulder line on the body (all three. to the left of the shadow) and their translational match on the head rim 1000 metres above (on the right). The head match is just showing the section of cove that matches to the most straightforward ‘s’ shape on the boulder line directly below it. Again, it’s difficult to depict and just representative here for the same reason as given for photo 4. However, photos 6 and 7 show that the same ‘s’ shape is indeed very apparent on the underside of the head rim. 

Photo 6- the ‘s’ shape on the head underside that matches to the ‘s’ shape on the boulder line

Photo 7- the rest of the head cove rim line extended from end to end. Again, the two dots at the left hand end are either side of the napkin that’s virtually edge-on in this view. The green section matches to its respective green locations on the boulder line and the cove seating on the body in the above photos. 


The match of the perimeter to the boulder line is strong evidence that it slid from the boulder line.

The boulder line moves into shadow at the bottom of the main photo for this post. The matching to the Seth perimeter remains faithful for another few hundred metres which would be about a quarter of the matched length here. It will either be added here in due course or get its own post. The boulder line in this section is blends with very low, solid outcrops and with the crack at the base of the neck. This of course has profound implications for the provenance of the crack. That’s because Aswan (the newly named Site A) used to run along the perimeter of the crack. We know this to be the case because the Seth perimeter (which includes Aswan) matches translationally to the boulder line which follows the crack. 

Either the crack was deep and dominant and caused the rift of Aswan from the crack line; or the rift was just the Aswan onion layer and caused the crack to go a little deeper at its bottom. This perhaps calls into question the theory that the crack is solely due to tensile stresses in the neck. It would be very susceptible to those stresses but its location is related to an onion layer tear, whether the tear was brought about by the cracking neck or vice versa. 


Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0
To view a copy of this licence please visit:


All dotted annotations by A. Cooper. 


67P/Churyumov-Gerasimenko. A Single Body That’s Been Stretched- Part 46


Unannotated version of header

Header reproduced with the outline of the depression (blue) and its lowest point (green). 


Narrative key for the header:

All three photos above are counted as photo 1, the header photo. 

The two lower orange lines in the header were described in Part 45. The uppermost line is new for this part. It’s the original seating of the orange crust. This was its position before it tore from the red/green crust and lurched 150 metres east (down in this view). It moved to the second line down, which is the classic seating in Part 45. It stopped here for an unspecified time before carrying on east. This period of stasis was long enough for the roundish features to form between it and the still-seated red/green slide. After continuing on eastwards, the crust eventually came to rest at the bottom line i.e. where it can be seen today. 

The blue line shows the perimeter of the depression where all the outgassing happened. The top of the frame can be considered as its top (western) edge which is actually just off-frame and parallel to the edge. The same applies on the right with the blue line turning parallel to the frame edge when a little way off-frame. The green dot is the lowest point of the depression and of the whole of Imhotep. 

The depression slopes up quite a way on its eastern side i.e. from the S-shaped blue line component at the bottom towards the lower orange line. The lower orange line is the top of the eastern rim. The eastern rim is the highest side of the notionally four-sided depression and is defined by the orange crust perimeter that slid. 

The other three sides (right, left and top) are lower and more defined so they define the edge of the tray-like depression more clearly. 

The bottom, eastern S-shape in blue runs almost at the bottom of the tray before it slopes up to the eastern rim. It’s slightly above the bottom though so as to be in keeping with the height of the rest of the tray rim. Despite being seemingly arbitrarily placed so as to satisfy this contour line requirement, it defines the eastern and southeastern edge of Henrik’s pancake which was carved out by the outgassing around the edge of the depression. The now cut-out pancake eventually slid south to its current position, next to and south of the depression (see the ‘Imhotep Sliding Crust Matches’ page in the menu bar). The S-shaped blue line therefore defines the eastern extremity of that outgassing. 

You may realise that the orange slide crust was in between the pancake and the depression and so should surely have prevented such carving. But it seems the orange crust slid out from under the pancake which then dropped down and was thereby carved out as a pancake. Prior to that it was just a large area of pastry drooping and sliding around (actually slowly southwards) waiting for the depression gas stencil to cut it out. There is much evidence for this dropping down after the orange slide had slid out from under it but it’s beyond the scope of this post. However, it’s telling that the pancake is not just the same shape as the depression at this eastern end (as it is at north, south and west) but is also turned up, looking like real pastry. The turn-up is the same as the slope it was sitting against. That is, the beginning of the long, eastern slope.


The reason the second orange line down was used for the seating in Part 45, instead of the top line original seating, is that the second line down is easier to match to the orange crust line at the bottom. It’s only when you become familiar with these two lines that you start to see the translated matching features on the top line. I didn’t see them at first, or at least, not in the sense that they matched well to the orange crust perimeter. However, I already knew where the line ran because it matches to the green slide crust perimeters that sat there. 

The exact positions of the red and green (mostly green) seatings in the depression have not been published as posts in their own right as of this post’s publication date but they are touched on here and towards the end of this post. This is because we need to keep introducing the context of the other slides going on around the orange slide in order to understand the overall sliding mechanism on Imhotep. But lower down, the orange slide will continue to be the main focus of this part. 

The Khepry component of the green slide is really a loose section of the red slide. The Khepry component is at the far west of Imhotep. It’s one of the three green slides (see the ‘coloured slide recap’ heading plus map, below). The green slides are not closely related in the mechanistic sense but are similar in that they involved layers that rode on the main slides but delaminated from them, sliding on further. The Khepry green slide slid with red and then slid on further when the red crust perimeter came to a halt on the other side of Imhotep. So Khepry green started in front (east) of red at their seating point and so was sitting mostly in the depression. It ended up behind (west) of the red perimeter when red came to rest. It followed the red slide vector as it travelled across the smooth terrain and slid at the same time as the red perimeter slid because it was under the same tensile forces. This is why lower down in this post the terms red, green and red/green are used interchangeably. Red is referred to when we’re thinking of the entire red slide process as described in Part 43 but bearing in mind that this little patch of green was really part of it for all intents and purposes. 

Khepry green is often referred to simply as ‘green’ because we’re a long way from even mentioning the other two green slides at this point. The Khepry green slide crust seating will get a lot more airing but for now it can be loosely defined as the area between the top orange line and the red slide seating perimeter. The red perimeter is running largely across the top of the frame in the header, along the top (western) edge of the depression. Alternatively, the green slide can be visualised in terms of the area of its seated crust. So it can be regarded as being most of the western half of the depression, west of (above) the top orange line. The fact it was contained almost entirely within the depression is an indication of why it was looser and more prone to slide on further. This will become clearer when the processes playing out in the depression are described below. And of course, now we’re looking at the depression perimeter in detail, we can see that the orange slide took up the other, eastern, half of the depression. The two perimeters clamped together along a line across the middle and that line is the top line in the header. 

The first part dealing with the orange slide was presented in Part 45. This part continues to focus on the orange slide but delves deeper into the mechanisms that might have induced the tearing of the orange crust, its sliding impetus and also the resulting phenomena brought about by the cracking open of the crust layers. The orange slide is intimately related to the slides around it so the next sub-heading is a short recap of the four coloured slides (see Part 42 for more detail. Part 42 is the Imhotep slide overview). The orange slide is also related to the paleo rotation plane and the depression, which hosts the roundish features so photos of these are shown below, along with the photo of the slides. 


Photo 2- the four coloured Imhotep slides (Part 42 header, reproduced). 

Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

Photo 3a- the orientation of the paleo rotation plane (key below).

The current rotation plane, which is today’s equator is in blue dots. The paleo rotation plane, which is the paleo equator, is in small brown dots. Please ignore the larger brown dots which are the former estimation of the paleo plane. It was estimated before the Imhotep paleo plane signatures were recognised for what they are. Hence the circa 3° adjustment (this photo is culled from the ‘Paleo Rotation Plane Adjustment’ page in the menu bar). The current plane and the paleo plane are angled at 12°-15° to each other. Only the red and orange slides are shown here so as to illustrate how their slide vectors are in opposite directions along the paleo rotation plane. 

Photo 3b- the outline of the depression, marked blue. This is the same perimeter as shown blue on the header photo. You can see it’s very nearly dead in the centre of the base of the body diamond shape, which is significant. The depression hosts the strips of roundish features and several curiously smooth, shiny ‘lakes’. The lakes are certainly solid even if, as this blog suggests, they were once free-flowing slurry. 

In Part 42 we looked at all the crust slides on Imhotep and divided them into four types: red, orange, blue and green. The terms “crust”, “crust layer” and “onion layer” are used interchangeably here. Crust carries with it the connotation of being somewhat devolatilised and also perhaps more susceptible to cracking and sliding, whereas onion layer reminds us of the concentric nature of the layers (each around 50-100 metres thick). The four coloured slides involve the first two layers and the second layer is sometimes called “the second crust layer” down to remind us that it’s the one under the surface crust and not just the second layer in any number of otherwise confusing crust piece combinations and mini delaminations.


Red slid 700 metres west. To be exact, it was in a direction 12°-15° south of today’s west direction but ‘west’ is definitely the operative word because this slide vector was exactly west down the paleo west direction at a time when the paleo plane held sway. The red crust slid from a tear line that ran along the short axis of the diamond-shaped body. The slide vector was at 90° to the straight tear and that is why the smooth terrain of Imhotep is uncannily rectangular. The sliding red crust perimeter ‘printed out’ the rectangle like the boom of a screen printer or a screed rail over wet concrete.


Orange slid 450 metres in the exact opposite direction to the red slide, 12°-15° north of east, and involved two separate tiers. Again it was exactly east along the paleo east direction. The fact that red and orange went in opposite directions is highly significant and implicates spin-up as the reason for the sliding behaviour. The first and most important component of the orange slide was shown in Part 45 and it will be the main focus of this part as well, while we get familiar with the finer points of the slide and its implications for the evolution of the roundish features in the depression. The reason for recapping the different slides in this sub-heading is that they are all interrelated and need to be at least partially understood at this stage so as to understand how they and the orange slide betray the evolution of the whole depression, not just the roundish features that are so closely related to the orange slide itself. 


Blue involved two sections as well. However, this wasn’t two tiers going in the same direction like the orange slide. It was two sections that originally kissed along the long axis of the diamond-shaped body and tore from each other. One slid north and the other one to the south. That means they both tore from the diamond’s long axis, which is hypothesised as being the paleo equator in this blog (see the next sub-heading). The paleo equator is quite close to the current equator that also runs longways across Imhotep.

The blue slide is the only slide that involved the top crust layer. Orange and red were both lower, second level, ‘onion layers’ of crust and originally sat under the blue top layer. That’s how we can have so many pieces tearing from each other and sliding out of such a small area as the depression. This is why the depression is a relatively deep hole and exhibits so many signs of catastrophic outgassing.

It should be restated here that the depression is almost exactly in the centre of the diamond-shaped base of the body lobe (see photo 3b). It straddles the long axis line and is set back east along it from its exact mid-point by a few hundred metres. If ever there were to be a rupture in the crust due to spin-up and on such a symmetrical shape as the diamond, the depression on the centre point would be the most likely point. It’s where all opposing centrifugal tensile forces are balanced and therefore where a tear right the way across the short axis would be most likely. The red slide tore from an almost straight line running exactly along the short axis and the depression sits at the centre of that tear line. This is why the depression is itself one of the 17 paleo rotation plane signatures. 

The depression is also where the longitudinal and Coriolis forces meet so there was a radial element to the overall force vectors. This was especially the case for some of the top crust sections that rode longitudinally on the main red/orange slides while delaminating and sliding north or south at the same time. This caused up to a 45° slide vector and is exhibited mostly by the green slides. The top orange tier exhibits it slightly as well and could at a pinch be designated as green. But it was predominantly eastwards with the bottom tier so it scraped in as orange. 

It is thus the overall radial element of the forces that probably explains why the depression is at the centre, small, circular and deep. This trait is over and above the long tear line that is so clear, running along the short axis. Moreover, the blue slide tore along the long axis and the two axes cross in the depression within metres of the very lowest gravitational point on the base. The depression is therefore the focus (start point) of all the sliding, whether translational or radial and it’s the kissing point of all the tearing crust pieces. No wonder the depression is sited here and is a site of furious outgassing. 


The green slide was designated for those sections of crust that were originally riding as a part of another slide. But they either continued on when the main slide stopped or travelled at 90° to the main slide while riding on it, resulting in a 45° effective travel vector. They are therefore sub-delaminations from the main slides. Green is a hybrid of top crust and second layer crust. The page entitled ‘Sliding in Western Imhotep’ (see menu bar) touches on one of the green slide components. It will eventually be expanded into a post because as it stands, it’s describing only the last part of its slide and not its seating in the depression. 


Photo 4 (which is photo 3a, reproduced).

The paleo equator is also the paleo rotation plane, sitting at 90° to the paleo rotation axis. Similarly, today’s equator is today’s rotation plane and it sits at 90° to the current rotation axis. The paleo plane was first described in Part 26 and crops up in most parts thereafter. There are seventeen symmetrical stretch signatures that betray the paleo plane. They straddle the plane along the paleo equator. The paleo equator is the line that the plane defines as it emerges at the surface of the comet. The seventeen signatures run round the entire comet, head lobe as well as body lobe, and are described at length in the ‘Paleo Rotation Plane Adjustment’ page in the menu bar. 

Today’s rotation plane runs at 12°-15° to the paleo rotation plane because the paleo plane precessed about the long axis of the entire comet (as opposed to the long axis of just the body diamond) by 12°-15° to arrive at today’s rotation plane orientation. 

The long axis of the whole comet is a line running through the centre of the comet from the centre of Apis to the centre-top of the Hatmehit cliff. So it runs through the body at an angle, then through the neck and finally through the head at an angle. Conversely, the long axis of just the body lobe diamond runs from tip to tip through the centre of the body lobe. It projects a line to the surface at Imhotep and that is the hypothesised paleo equator. Although this is a distinction between two lines (the comet’s internal long axis line going through both head and body and the derived paleo equator line on Imhotep) they both sit in the same plane which is the paleo rotation plane. 

The fact that these two lines both sit in the paleo plane should be no surprise since, for the rotation plane to have precessed about the comet’s long axis, from paleo to current, that long axis line was the only line that was residing within the precessing plane continuously throughout the precessing process. So the long-axis line had to be in the same plane as the paleo equator across Imhotep at the start of the precessing operation. By the same token, the comet’s long axis line is today still in the same plane as the paleo plane, and the paleo equator across Imhotep, and therefore the paleo equator all round the comet. Similarly, it is in the same plane as the current rotation plane and current equator all round the comet. The comet’s long axis pulls off this neat, double-plane-affiliation trick by being the one and only single line that resides in both planes. It owes its unique status to the fact that it’s the axis of precession between the two planes. And since the two planes draw their respective equators on the comet’s surface wherever they emerge at the surface, they consequently cross each other at Apis and the Hatmehit cliff. That is, at the two ends of the comet’s long axis. They cross at 12°-15° to each other, which is the angle between the two planes. Moreover, they cross only at these two points because they are the two points that performed as the gimbals during the precession process. If we could dissolve the comet away suddenly and inspect just the two planes in their entirety, they would be seen to be intersecting all the way along the long axis of the comet and nowhere else. 

The red slide (Part 43) and the main orange slide component described in this part had slide vectors running exactly along or parallel to the paleo equator line. We know this because the translated matches haven’t rotated or drifted away from this slide vector: when you draw a line between each pair of mini matches along the two matching perimeters, crust-to-seating, for either slide, that line is the slide vector for that mini match from its seating. For the red and orange slide, all such slide vectors run parallel to each other and parallel to the paleo equator. They are therefore within planes that rotate parallel to the main paleo rotation plane that defines the equator. This strongly implies that each and every mini match along both slid crust perimeters was intimately related to the paleo rotation plane and therefore to the rotational behaviour of the comet. But the mini matches along the two slid crust perimeters actually define those perimeters. So it follows that the slide behaviour of the entire red perimeter and orange perimeter was also closely related to the rotational behaviour of the comet. 

It should be added that the slide distances of all the mini match slides are the same or very nearly the same, showing that the translational movements of the red and orange crust perimeters didn’t suffer any significant rotation superimposed on the pure translations. That shows that the centrifugal forces remained the same at all places and all mini matches along along both the red and the orange sliding perimeters. This is a proxy signature for the parallel plane signature because the original tear line was parallel to and equidistant from the rotation axis of the comet (Imhotep being flat). Therefore, all mini matches sitting in each parallel rotation plane had the same rotation radius and should have experienced exactly the same force at any one time during the slide. This translates to them all experiencing the same overall force over the whole slide. Thus, the perimeter line of the slide had no reason to rotate due to any anomalous force vector (torque) at one end. The translational symmetries of the crust perimeters bear this out perfectly, showing almost no rotation. 

If the the perimeter translations had suffered torque forces, the three lines in the header wouldn’t be so stunningly parallel. You’d expect some rotation due to frictional anomalies and the second line is slightly clockwise but with all these crust layers being near-weightless on spin-up, the overall friction would have been significantly reduced, let alone the anomalies. In fact it’s worth bearing weightlessness in mind when constantly stumbling upon all these surprisingly neat slide patterns and symmetrical shapes as well as lower layers sliding out from under upper layers. Weightlessness leads to much-reduced friction between layers and it’s frictional anomalies that always introduces anomalous torque. If they didn’t, farmers would use harrows with a single a central mounting, not the obviously practical end mountings. 

So these two slides, red and orange, were almost certainly brought about by the centrifugal force vector of the rotating comet. The sliding would have been happening as the single-body comet was stretching and before the head lobe finally sheared and rose on the growing neck. For post-head-shear matches and behaviour see Parts 1-8, 17, 21, 24, 25. The enhanced centrifugal force vector would be due to spin-up to a 2- to 3- hour rotation period (see the ‘Spin-up Calcs’ page in the menu bar). The spin-up torque would have been provided by asymmetrical outgassing which is known to cause spin-up or spin-down, to a greater or lesser degree, on all active comets. 


Whilst this part will soon focus predominantly on the orange slide, it’s presented with a lot of context as to what was happening around it in the form of the red and green slides and above it in the form of the blue slide. Otherwise it’s just an isolated match and slide with no context. The context is built upon and elaborated as we progress through to the blue and green slides in future parts. 

There are no fewer than twelve pieces of crust that have been identified which kiss and/or nest together in the depression when the ‘stretch movie’ or ‘slide movie’ is reversed. In this sense they are highly interrelated. Although it’s impossible to explain them all in one go, it’s as well to bear in mind from time to time that the orange slide had other crust layers sliding past it and over it (most likely near-weightless crust so little friction) while it slid along its own slide vector. Furthermore, it tore away from the green crust pieces it was originally kissing in the depression. It would be a good thing to try and remember the references to the surrounding slides as they are mentioned so as to become gradually familiar with their direction, crust layer order (above or below other layers) and their crust-piece neighbours. It’s like a detective story where recurring instances of matches and slide directions keep reinforcing each other, building the overall narrative. Each casual reference to a slide will eventually be elaborated on in a fully detailed post for that slide. But if you’ve remembered its behaviour, you’ll be one step ahead when it gets related back to the orange slide and other casual slide references in this post. 


As is stated from time to time on this blog, the processes and mechanisms are stated here as fact to save on laborious use of the conditional tense and endless caveats and qualifications. They are hypothesised processes and mechanisms, although some are nevertheless nearer to certain than others. Hence, the orange crust slide is very compelling indeed, as shown in the header. The mechanism described below for the appearance of the roundish features in the depression is slightly less compelling due to being one stage further removed in the chronology. However, the reader is currently at a disadvantage in assessing the validity of this mechanism because it hinges on how well the crust sections fit together in the depression when the slides are reversed. The fits have all been identified but it will take several more parts to present them. 


The orange crust slide in Part 45 and elaborated on in this part is the lower tier of the orange slide. The upper tier (see the Part 42 header, above) will be presented later. The upper tier sat on top of, and to the north, of the lower tier. North would be to the right and out of frame in the header. The upper orange tier slid further on eastwards and has been matched both to the lower orange tier and to the depression (not yet published as of the date of this post).

The lower tier is more important because its rupture from the depression was very likely the cause of the icy patches along its edge. Its seating line runs faithfully along the perimeter of the roundish features in the depression. These are the roundish features described in Auger et al. (2015). They were posited to be ancient gas dykes in that paper. They make the depression look like a dried-up water hole or extinct geyser field. 

This blog hypothesises that the roundish features were formed by slurry, thereby implying liquids. This is consistent with all the slurry signatures around the comet so far identified, starting in Part 5 back in December 2014 and built upon over 17 months and another 41 parts. 


Although slurry is implicated as a possible concomitant to stretch and head shear, stretch theory doesn’t rely on the presence of slurry at all. Neither does slurry rely on stretch theory. However, both theories reinforce each other: slurry signatures abound in all the key areas that stretch dictates as being locations of head shear or crust cracking, thus reinforcing the stretch narrative. This is because the stretch evidence depended only on matches. It was only after the matching was done (head-to-body for head shear and crust to seating for crust slides) that the slurry signatures were noticed (especially in Parts 7, 17 and 21). No matching has ever been instigated based on a slurry signature, although nowadays, such signatures are expected to arise at times along the matched perimeter, acting as corroboration. The finger on the second seating of the orange slide is a case in point (topmost feature of the middle line in the header). It’s entirely formed from flash-frozen slurry and matches to the actual crust finger that sat on it and allowed it to take on that shape. The reasoning for it being slurry is given further down. 

And as far as stretch theory reinforcing slurry theory is concerned, all those slurry locations had to be originally tens to hundreds of metres below the surface according to the tearing and sliding behaviour. And that means they were more amenable to being under sealed, lightly pressurised layers. That’s the only way they could harbour liquids. Conversely, no areas that were on or near the pre-stretch surface have slurry signatures, or at least, any easily identifiable ones. Those areas are: Ash, Khepry, Apis, lower (east) Aten upper (west) Bastet, upper Ma’at and the Cliffs of Aten (specifically the cliffs’ ‘whiter’ tops in Babi).


In the case of the depression, the slurry would appear to have found its way in between the layers of crust, specifically, between the second and third layer down. It then outgassed rather furiously when the second layer down cracked apart and slid open in exact opposite directions (the red/green and orange crust slides). This furious outgassing caused the roundish features. The slide vectors were along the direction of the paleo equator line, red going exactly west and orange going exactly east. Not 12°-15° off west and east, as briefly mentioned above, because we’re now describing the slides with reference to the paleo rotation plane and its equator that held sway during spin-up before head shear. Above, it was with reference to the current equator which is at 12°-15° to the paleo equator.

The roundish features in the depression therefore betray the crack between the red/green and orange crust components. It’s about 150 metres wide at the left hand (southern) end and 200 metres wide at the right hand (northern) end. We’ll just say it’s 150 metres wide for shorthand from now on. It opened up and remained at 150 metres width for an unspecified period of time before the two slides continued on their way for good. The cracking open allowed sudden outgassing to occur from below that respective layer, the second crust layer or onion layer down. The hiatus at 150 metres in width then allowed that outgassing to be channelled through this comparatively narrow strip. The strip’s signature is betrayed by the strip of roundish features seen today and hence the crack’s signature is also betrayed by the roundish features. This crack width is shown in the header and it’s reproduced here:

Header (reproduced).

The original seating of the orange crust is the uppermost line. This was its position before it tore from the red/green crust and lurched 150 metres east (down in this view). It moved to the second line down, which is the classic seating in Part 45. It stopped here for an unspecified time before carrying on east. This period of stasis was long enough for the roundish features to form between it and the red/green crust, which was still seated. 

You may notice that there aren’t any obvious roundish features at the right hand end of the crack, the northern end. So this sub-heading isn’t just about how they formed but also why they didn’t form in certain places. There’s just a smooth area with some some roundish features around its edge at this end of the crack. This smooth area is biased towards the upper line with some roundish features creeping along the lower line. This might appear to upset the theory that opening up the crack allowed the outgassing to commence and cause the roundish features to appear. However, there are sections of sliding crust whose shapes and slide tracks have been matched to this area. Not just the area in general but two shapes imprinted on the slurry and defining 90% of the area and its perimeter. They correspond to two other slides that remained for a while before sliding in their respective directions. 

There is also no evidence of roundish features at the other end of the crack. This area didn’t have any lingering slide pieces staying put, clamped against the upwelling slurry and suppressing roundish features. But it has to be remembered that the pancake was still in place, sitting above the orange slide and acting as a lid or ceiling some 80-100 metres over the floor of the crack. The orange slide had shunted east while the pancake stayed largely in place but the pancake did in fact shunt south while keeping this left hand end of the crack well covered in the process (see photo 6). 

Regarding the lack of roundish features at this left hand (southern) end of the crack, the orange slide crust itself seems to have acted as shuttering in a stop-start sequence of slide shunts which left longitudinal voids. Those voids appear to have filled straight away with slurry, the orange crust acting as shuttering on one side and the hardened wall of old slurry as shuttering the opposite side. This mechanism can’t be as clear or likely as other mechanisms in this post but it would explain the terracing elegantly. Notice how each terrace is slightly bulbous across its width, as if it overfilled a little perhaps and froze over in that profile once the gases in the slurry had been spent. You can see four terraces between the two lines defining the crack at this end. There may be a fifth above the purported original seating line so the line may be in need of shunting up one terrace just in this short stretch. It’s not known at this stage why the crack opening promoted terraces and not lines of roundish features at this end. However, there are lines of small roundish features along the terraces that are further to the west. 

If you scrutinise the unannotated version, you can see a slide track where the vertical part of the top line tracks down to that very large roundish feature. The top half of this line down to the roundish feature and then across its top is also the seating line for a section of the green slide on the other side of the Imhotep smooth area. They’re the same shape and a spidery line runs all the way across the smooth terrain from the that top sharp point to the southern edge of the green match. The spidery line was a bonus. There was plenty more evidence for it before the line was discovered. 

It seems that the same stop-start terracing process happened in the opposite direction on the other side of this common seating line for the orange and green slides which slid in exact opposite directions. You can see that at the point of crossing the orange line, the terrace lines stagger, showing that a different stop-start shunt event was occurring with a different piece of shuttering (the green slide crust) deciding where it wanted to stop and start. But the process was identical, shunting up that vertical portion of the orange line instead of downwards for the orange slide. This paragraph concludes the reasons why roundish features didn’t occur at both ends of the crack. We can now continue with the roundish feature formation mechanism.

The outgassing in the new 150-metre crack would have occurred because of the drop in pressure under the second layer where the slurry was. The pressure would have to have originally been at least 611 pascals (the triple-point pressure value for water). Hydrostatic pressure from the weight of the two layers of crust above, circa 180 metres thick, would furnish only 20-30 pascals, even for a non-spinning comet, and much less for one that’s spinning at the required rate for stretch and sliding crust. For a 180-metre column of one metre squared:

 180m x 530kg/m^3 x 2.9 mm/sec^2 = 27.66 pascals

GM/r^2 @ 1500m radius = 667.4/ 1500^2 = ~2.9 mm/sec^2, where G is the gravitational constant and mass, M is 1E13 kg. 

1500 m radius from centre of a notional sphere for Imhotep is estimated for the unstretched comet. In reality it would be partially stretched and potato shaped, reducing the acceleration due to gravity a tad at Imhotep to a little below the 2.9mm/sec^2. This value reduces progressively on spin-up. 

This calculation gives nowhere near 611 pa and it implies that the underside of the layer that cracked was pressure-sealed somehow by refractory material. Though this may be difficult to envisage, it behoves us to entertain the possibility because the evidence from the orange slide and its slide chronology bears out the existence of slurry and slurry requires pressure. 611 pascals is 0.006 atmospheres so it would be only lightly pressurised. 

The sudden drop in pressure from 611 pa or more, when the crack opened, would have instigated the outgassing. The outgassing was via the liquids in the slurry flashing to gas and causing the roundish features as it forced its way out. The roundish features were flash-frozen where they formed when the liquids were eventually spent. At that point, only the refractory silt in the slurry was left but it’s described as being flash-frozen because the silt assumed the shape of the roundish features as they were when expelling their very last supplies of gas. And that gas was coming from the very last supplies of liquid below the surface and within the slurry contained in the roundish features. 

It’s becoming increasingly likely that the liquid was water because several papers have implicated water as being responsible for the visible surface ice on Imhotep. They have done so via different methods: Auger et al (2015) via the detection of non-devolatilised boulders; Groussin et al. (2015) via blue reflectance using OSIRIS filters (though still admitting the possibility of CO2); and Filachionne et al. (2015) using the VIRTIS instrument on the Rosetta orbiter. 

To be clear, no Rosetta scientist nor any scientific paper is supporting the idea of liquid water or any other liquid on 67P, either today or at any time in the past. And indeed, all the supposed water ice signatures at the surface today are just that, ice. The reason for this blog invoking liquids is the clear slurry signatures that behove us to test scenarios regarding a warmer, lightly pressurised interior that would support liquids. This might have occurred for long periods or just for a short time during spin-up and stretch. 

One important point is that what is seen as surface ice today on Imhotep had to be buried very far below the old surface before the four coloured slides parted like the Red Sea to reveal it. Although it’s assumed by cometary scientists that ice does exist hundreds of metres below the surface and right down to the centre, it’s only inferred from the comet’s density and from cometary models. Since the crust slides on Imhotep amount to a former thickness of some 100-200 metres of crust sitting above today’s ‘surface’ ice signatures, it follows that at least water ice definitely did exist at that depth before the slides occurred and likely still does elsewhere where slides didn’t occur. That’s if you subscribe to the crust-sliding theory in the first place. This wouldn’t come as a surprise to the scientists but the point being made here is that it’s definitive proof rather than inferred or modelled. And if water existed at this depth (if the slurry signatures are to be believed) it wasn’t ice but liquid. 

Liquids could be the reason the smooth area (that large rectangular screed) looks like hardened, devolatilised slurry instead of dust. From a distance at least. Up close, at the 30cm/pixel scale, it’s grainy so it’s more like a very big, flat expanse of dumped bricks, rather large ones at that. It really doesn’t look like fine dust as is often supposed and, as Auger et al. 2015 point out, the large 20- to 40-metre boulders on this area appear to have little or no dust deposit on them, which we would surely expect to be the case if they were surrounded by a dust field. Moreover, Groussin et al. 2015 witnessed a massive slump on 40% of the smooth terrain just before perihelion in 2015. The slump was five metres deep and implied a substantial erosion event- well over a million cubic metres. This would have involved a large loss of volatiles via outgassing and yet there was no evidence at all of dust accompanying that outgassing coming through the smooth terrain. It’s as if there’s little or no dust there to speak of and the gas was simply passing through the bricks. If it is indeed hardened slurry, the bricks are the non-volatile, refractory leftovers, the silt if you like. Their brick size could be explained both by the nature of the sudden devolatilisation (how fast and violent it was) and by cosmic sintering as well as thermal stresses. So you wouldn’t expect it to appear smooth like hardened ice at high resolution. However, from a distance it betrays the characteristics of the liquid it is purported to have once been. It’s even glooping over into the depression in folds- caught in mid-gloop as the folds froze on devolatilising. 

When outgassing, the roundish features in the depression would have been like mud geysers but the process would have involved brute-force gas pressure from BLEVEing liquids (flashing to gas) as outlined in Part 42. The process would not involve buoyancy like some geysers on the Earth because the gravity is too low. 

The well-defined circular walls of the roundish features are explained quite well by this theory of the liquids in the slurry flashing to gas. They would have hardened prematurely as new gas welled up from the centre of the circle, possibly causing a circular dome in the slurry. Whether the dome burst and reformed or whether it remained in place while gas passed through it, the tendency would be for thicker, volatile-poor slurry to be forced to the circular perimeter as it gave up its last remaining gases and hardened fully on the perimeter. Hence the apparently hard, thick circular perimeters formed of refractory, devolatilised slurry. This would suggest a premature hardening at the rims before the centres hardened at the point of being spent of gas. 

The above theory invoking liquids explains the four different types of roundish feature observed in Auger et al. (2015): sunken, level, bulbous and smudged. The first three are various stages of slurry-filling of the solid cylindrical walls belonging to the roundish features. The fourth corresponds to roundish features that were both very stilted by being under crust and also smudged by that crust sliding over them. Witness the flattened area mentioned above, which was said to have two shapes impressed on it and which may have some stunted circles though it’s hard to tell and they’re not the best examples. 


One very faithful match in the orange slide is a roundish feature near the left hand (southern) end in the header photo. It’s a ’roundish’ feature by classification but is in fact polygonal. It’s just to the right of and below the finger match in both lines (that is, the two lower lines in the header which are the classic Part 45 crust and seating). The roundish feature is polygonal because its solidifying perimeter wasn’t allowed to assume a circle like the others via equilibration of radial forces. It was too close to the edge of the crack and so its wall was moulded against whatever shape the orange crust had. That happened to be polygonal. The polygonal ’roundish’ feature matches very well to its polygonal mould on the crust 450 metres away. This behaviour is entirely consistent with the presence of slurry during the formation of the roundish features:

Photo 5- the two matching polygonal features (in fine detail so full zoom is required). This is cropped from the header.

Orange- the classic orange crust perimeter (bottom) and seating perimeter (top). Also an arc midway between the two polygons which appears to be a mini delamination where the rounder support structure of the polygonal crater rim got left behind. 

Green- these are in two pairs, one pair at the top and one at the bottom. One pair denotes either end of the polygon and the other pair, either end of its seating. The top left green dot is beyond the orange line because the orange line follows the supposed slurry line that formed under the finger. The side of the finger presumably extended slightly further than the slurry. Seeing as it’s the same shape as its rocky match that sat on it, it’s virtually impossible to think up a dust deposition scenario that would satisfactorily explain this phenomenon. How could dust arrange itself in the shape of the finger that had once sat on that spot and do so only after the finger had slid away? So this is the second of the two independent tell-tale slurry signatures. Even though it’s right next to the other signature, they are independent proofs because they involve completely different processes. One is invoking the formation of a shape in a mould which implies slurry and the other invokes slurry sitting protected beneath the finger before it slid east and neatly out of the shape without disturbing it. 

Red-the slide track for the polygonal feature. The slide track appears to continue on downwards past the end of the red line, possibly all the way to the crust perimeter match, but it’s left unannotated because it’s less obvious and somewhat debatable. 


Photo 6- The bled matches. 
Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

White- slide track

Green and dark blue- the original seating (on the right) and (green only) the point where these green sections of the perimeter came to rest before the bleed event occurred. 

Red, yellow, mauve and orange- corresponding features, in the depression (on the right), and bled through to the surface of the pancake (on the left). 

The crust sections didn’t slide away immediately, presumably because there was top crust sitting on top of them in the form of Henrik’s pancake (Part 42) which is the main southern component of the blue slide. Henrik’s pancake retains bled matches on its surface from the roundish features outgassing in the depression. It’s quite remarkable that these should remain discernible after travelling up through 80 metres of crust but they are undeniably there. Photo 6 is also now on the ‘Imhotep Crust Sliding’ page. It shows that Henrik’s pancake stayed in place while the crust below it cracked and the roundish features formed. 

The bled features on the surface of the pancake are congruent with the lines of roundish features that used to be directly underneath those bled features. So the cracks would have been underground, longitudinal chasms, 150 metres wide and 80 to 100 metres deep. They would have had 20- to 70-metre-wide ‘geysers’ spouting gas on their floors. Some of that gas (possibly along with some attendant refractory material) bled through the 70-80% porous pancake to the surface. The gas escaped to the vacuum of space at the surface. In doing so, it either modified the surface of the pancake via a mechanical mechanism or it left behind the refractory material on the surface as a precipitate. Either way, the signature lines of the cracks hosting the roundish features were imprinted on the surface of the pancake as the gases passed through from the cracks below. The faithful, bled matching of the lines suggests that the pancake sat right against the top edges of the cracks, forming a roof to the underground chasm. That would mean the gasses and any attendant slurry had to escape through the portions of the pancake that were directly above the cracks, hence the faithfulness of the bled matches. 

The furious activity below Henrik’s pancake is the very reason for the pancake’s existence. It fits perfectly to the depression and its turned-up ‘pastry rim’ was very likely torn and pushed up by the gas escaping around the edges of the depression. Put another way, the pancake was the lid on the depression and its remarkably similar shape was literally carved out by the activity beneath it and emerging around the edges of the depression. The pancake match is on the ‘Imhotep Sliding Crust Matches’ page as well. It’s been very slightly refined since posting and will be updated in due course. The current match is to the 15-metre scale in places.


Before the crack opened to allow the roundish features to form, the orange slide crust was attached to sections of crust from the red and green slides. These red and green sections have all been identified (Parts 42 and 43) and they fit into the depression in a perfect puzzle-fit. All the crust perimeters match to the perimeters of the meandering roundish feature bands. The crust sections therefore fit to each other because the roundish features are simply the signatures of the fissures between the sections after they cracked apart. The orange slide is just one example of a fit to the roundish features and the others will follow in the future parts on the orange slide as well as the green slide (and a revisited red slide). 

Since the strips of roundish features are hypothesised to be the widths of the cracks, it means that the lower tier of the orange slide in this post and Part 45 should fit not only to the bottom (eastern) side of the roundish features but also to the other (western) side. And indeed it does:

Photo 7 (header reproduced a second time)- the orange slide fits to the other side of the strip of roundish features as well.

The upper orange line is the original seating of the orange crust on the other side of the roundish features and kissing the red/green slide perimeter along its length (exact red/green seating is not shown nor yet published as of the date of this post).

There’s a small gap at the left in this new, upper line where there is no discernible seating line to trace it along. The rest of the line does predominantly follow actual lines including those implied by shadowing, most noticeably for the dip in the middle. There’s a section just to the right of the dip that’s also not well-defined but fairly constrained by the lines either side. The gap mentioned above was left open because there’s no line at all and the lines either side give no indication as to the extent of this much smaller dip. Notice that, whatever its shape, it’s concertinaed together more at this point than the classic Part 45 seating below it and on the other side of the roundish features (see photo 8 for an explanation of this). You need to check the unannotated version to verify all these statements. 

Since the orange slide was originally attached seamlessly to components of the red and green slides along this original seating line before it tore away from them, it implies that the red and green slide components nested in some configuration so as to run faithfully along this upper, (western) line as well. We shall come to see in future parts, that they indeed did so. 

Photo 8- the red slide tracks betraying the concertinaing and the stepped seatings that caused the terracing. 

Photo 8 explains the anomaly towards the left of the line. This refers to the terracing process described above but focuses on the way the orange crust opened laterally as it slid east in that stop-start fashion. It thereby imprinted this behaviour pattern on the terraces themselves. 

The top line at this southern end is concertinaed more together. You can now see the red slide tracks. The right hand track follows the conventional downward (eastward) slide. The left track runs parallel to the right hand one until the last minute when it lurches to the left as well as continuing on down (southeast). You can actually see a former seating of the pointed section towards the end of this red track line. It’s just to the right of the currently marked seating and of course, the red track passes through it. The actual slid crust is 450 metres east at the bottom of the picture so we’re talking about different seatings which were stepped and predominantly from top to bottom (west to east) except for this one obvious lurch to the southeast. The stepped seatings are actually the raison d’être for the terraces we see today. This stop-start behaviour to the orange slide and consequent terracing will be studied in greater detail in the future. 

And finally, the blue slide was one layer above the red and orange slides and so it didn’t play a direct role in opening up the cracks for the roundish features. However, Henrik’s pancake, part of the blue slide, must have regulated the process of cracking below it and, to a small extent, the pressure on the liquids in the slurry. The green slide incorporates some top crust as well as second-layer crust that went quite a way under Henrik’s pancake. It extends so far as to nuzzle its way into that distinct curve in the middle of the orange line. It’s the large finger (another much larger finger, cited as one of the paleo plane signatures). It’s on the other side of Imhotep. It’s close to that other green slide section that abutted the terracing and was said to be traceable via the spidery line. However they’re not to be found in quite the same relational configuration where they sit today. The finger slid back a bit further. Both these green sections will get their own post. They were just mentioned so as to start to get a feel for what was happening on the other side of the crack. 


There will be one more part for the lower tier of the orange slide. It will focus on the large boulder just below the orange slide’s crust perimeter and its seating in the depression. It will also show more red slide tracks. However, we may visit the green finger mentioned above first because it helps better to start building up a picture of the puzzle-fit in the depression. 


Marco Parigi has a stretch theory blog as well. It describes many of the aspects of stretch in more concise terms than here and with annotated photos. It then links to the relevant posts in this blog for those readers who want the full, in-depth analysis:


Marco thought of stretch theory (including liquids) as it would be applied to comets in general, and did so long before Rosetta arrived at 67P-CG. My first realisation regarding the possibility of stretch was on seeing the first published close-up of 67P on August 6th 2014. 



Copyright ESA/Rosetta/NAVCAM – CC BY-SA IGO 3.0

To view a copy of this licence please visit:


All dotted annotations by A.Cooper