Michael Reed

Profession	Dermatologist. The documents read for this page call him “Dr. Michael Reed”12, “Dr. Reed”34, and, on two card records, “ML Reed MD”56. One sheet gives his affiliation and his field in the same clause: “dermatologist, Dr. Michael Reed at NYU”, followed by a description of him as a hair specialist1
How the record describes his work	”hair specialist”, in three separate messages by two writers172
Dates carried by the documents read for this page	16 December 2014 to 12 January 2015, on the New York clock. Seventeen of the nineteen documents are single messages from a span of sixteen calendar days in December 2014, 16 to 31 December, which is fifteen days elapsed; the eighteenth is a card-statement compilation carrying a transaction dated 30 December 2014, and the nineteenth is a household expense ledger whose own Date column books the charge on 12 January 201556
Relationship to Epstein	None recorded. He is named in Epstein’s office correspondence as the doctor another doctor recommended, an appointment Epstein’s office booked and paid for, and a merchant descriptor on a card record. Nothing read for this page records him writing to Epstein, speaking to him, meeting him, or acting at his request135
Acts by him in the record	None found in any document read for this page. His practice is twice a payee, and two messages record contact about the appointment, one of them naming his office, but no document read for this page was written or signed by him3568

Michael Reed is a New York dermatologist named in files released under the Epstein Files Transparency Act. In December 2014 another doctor recommended him to Jeffrey Epstein’s office as a hair specialist; Epstein approved the suggestion twice in one word each time, his assistant booked a consultation for a third person, and Epstein’s office paid the fee on his card. Nineteen documents read for this page carry that sequence. In every one of them Reed is a name other people write down.

A dermatologist who treats a patient has done nothing wrong. Nothing on this page is a finding of wrongdoing, and appearing in the released files is not evidence of it. Many people in these documents did nothing wrong. This page exists because the editor’s standing direction is that every doctor in the files should have an article, and what follows is the whole of what the documents record.

This page reproduces no clinical detail about any individual, and no address or telephone number for anyone. The person the appointment was made for is covered by the producing party’s bars in every document read for this page, and nothing is supplied for that person here: no name, no age and no circumstances, beyond the fact, printed on the face of the documents, that an appointment with a dermatologist was booked and paid for. Nothing about why it was booked appears in any document read for this page. The identity numbers, addresses, telephone numbers, and medical and family details of every person in these files are outside what this site reproduces, and that applies to patients and to doctors alike.

How the dates on this page are read

Every date and clock time below comes from a stamp, and the stamps are of two kinds. A stamp that carries a zone is directly comparable. A stamp with no zone is not, and a date may be taken from one only after checking what the conversion does to it.

New York was on Eastern Standard Time, five hours behind coordinated universal time, throughout the period covered by this page. At five hours, the window in which a universal-time stamp converts back onto the previous day is 00:00 to 04:59.

The production’s own Sent: headers on these sheets carry no offset, and on this thread they are coordinated universal time. That is not assumed here; it is established twice, from the documents themselves, by matching a message whose stamp states no offset against a twin of that same message whose stamp states one. Epstein’s one-word reply of 18 December is stamped 3:09:52 PM with no offset in one production and quoted in another as having been sent at 10:09 AM, a difference of exactly five hours.⁹ His second one-word reply the same day is stamped 8:03:06 PM with no offset in one production and 3:03:06 PM EST in another, again exactly five hours.¹⁰¹¹ A third check agrees from a different direction: a calendar reminder whose event time the sheet itself labels Eastern Time fires ten minutes after the notification whose produced stamp announces it.¹²

One stamp on this page stands inside the rollover window and its date does move. Epstein’s first approval is stamped Wednesday, December 17, 2014 1:12 AM. Converted, it is 8:12 in the evening of 16 December, and the thread it answers was written earlier that same evening.¹³ This page gives 16 December, and says so here rather than leaving a reader to find the discrepancy against the sheet.

A third clock appears on these sheets and this page does not rely on it. Where a later message quotes an earlier one, the quoting software prints its own rendering of the earlier stamp, and on this thread two of those renderings run one hour ahead of the universal-time conversion. Groff’s message of 16 December is produced-stamped 11:06 PM, which converts to 6:06 PM, and is quoted in a reply as 7:06 PM.¹¹³ Her message of 18 December is produced-stamped 7:04:26 PM, which converts to 2:04:26 PM, and is quoted as 3:04 PM.⁷¹⁰ The cause is not established here and is not guessed at. What matters for the sequence is that only the converted reading places that message before the zoned reply it provoked; the quoted rendering would place it one minute after, which cannot be right.

Each footnote below states which kind of stamp it relies on and prints the subtraction.

16 December 2014: the recommendation

On the evening of 16 December a message reached Epstein under the subject line “Hair Specialist for ▮ from Dr. Merrell”. Its sender is covered by the producing party’s bar and nothing is supplied for it here. The body is two sentences:

“Dr. Merrell is recommending dermatologist, Dr. Michael Reed at NYU”

The sentence continues by calling him a hair specialist and asking whether an appointment should be made. The apostrophe in that clause is destroyed by the production’s own transfer encoding, and the damaged form is not reproduced here. Below it the sheet carries a web address built from his name, which no longer resolves.¹

Woodson Merrell, the doctor named as making the recommendation, has a page on this wiki. The documents read for this page record only that his name was given as the source of the suggestion. Nothing read for this page records him speaking to Reed, writing to him, or knowing what became of it.

Epstein answered in the small hours of the production’s clock and the same evening of the New York one. His whole reply is one word, in lower case:

“yes”

It went to Lesley Groff, under the same subject line.¹³

18 December 2014: the price, and a second approval

Two days later the request reached the office again, and the exchange that follows is the densest material on this page. Groff asked Epstein:⁷

“Did you want me to make an appt for ▮ with the hair specialist Dr. =errell suggested: Dr. Michael Reed…please advise.”

Epstein’s whole answer is again one word:⁹

“Yes”

Within six minutes of that approval Groff wrote to the person the appointment was for:

”…I have the name of a hair specialist from Dr. Merrell…Jeffrey would like you to go and see him…I am happy to make the appt if you give me several day/times you are available…this time of year is always difficult!”

and gave his name and the same web address.²

Later that afternoon Groff reported back to Epstein with the fee and a term of business, and that message carries the only statement in the documents read for this page about how the practice operates. Groff had called to make the appointment, though the message does not say whom she called; the consultation on its own was $725.00, and Reed took no insurance. She asked Epstein to advise whether she should make it.³

Epstein replied, zoned this time, at 3:03:06 PM Eastern Standard Time. His whole message is two letters:¹⁰¹¹

“ok”

That evening Groff wrote to the person the appointment was for, copying Epstein and Bella Klein, under the subject line “Dr. Michael Reed”:

“…you have an appointment with Dr. Michael Reed at 2:30pm on =ues. Dec. 30th. I will call and pay for your consultation morning =f 30th. Please confirm back to me”

The two equals signs in that passage are printed on the sheet in ink, where the production’s transfer encoding has replaced the first letter of each word, and they are reproduced as the sheet prints them rather than silently repaired. The practice’s address and telephone number are covered on the sheet and are not supplied here.¹⁴

An hour later a reply came back to Groff, copied to Epstein and Klein.¹⁵

30 December 2014: the payment, and the appointment

On the morning of 30 December a reminder from Google Calendar fired on the office calendar. Its title names the doctor, the time of the appointment and the fee, and the sheet labels its event time Eastern Time:

“Pay for ▮ appt with Dr. Michael Reed today, Dec. 30, at 2:30…call ▮ $725.00”

The telephone number the title tells the reader to call is covered on the sheet and is not supplied here.¹²

Twelve minutes later Groff wrote to Bella Klein:¹⁰

“I am paying for this appt today on JE’s amex….will send receipt once I receive…”

A minute after that she wrote the reminder itself, to the person the appointment was for, copying Klein:⁴

“…just wanted to remind you of your doctor appt with Dr. Michael Reed today at 2:30. Bella, I have paid for the consultation on Je’s amex ($725.00)”

The reply records that the doctor’s office had been contacted and the appointment confirmed, and Groff’s answer to it is one word, “Great!”⁸

One further message on the same thread is among the nineteen documents read for this page; it reproduces the 30 December exchange and records nothing further about Reed.¹⁶

31 December 2014: a receipt, and what it does and does not establish

The next morning Groff forwarded to Bella Klein a notification of an inbound fax received the previous day, under a subject line of her own composition:¹⁷

”▮ Dr. Reed Visit receipt”

Three things about that document are worth stating precisely, because the page’s strongest-looking claim sits on it.

First, the sender of the fax is covered by the producing party’s bar in the notification panel, and nothing is supplied for it here. The subject line attributing it to Dr. Reed is Groff’s own wording, not the document’s self-description.

Second, the fax itself is not in the release as far as this page’s searches reach. What is in the release is a notification that a fax of two pages arrived, and the attachment named on the sheet is not produced with it.

Third, the two clock readings the sheet gives for that fax are three hours apart and only one of them carries a zone. The notification’s own header is zoned at 4:00:53 PM Eastern Standard Time. The panel inside it gives a receipt time of 01:00 PM with no zone stated for that field. On the unzoned reading the receipt would predate the 2:30 appointment by an hour and a half; on the zoned reading it follows it by an hour and a half. The sheet does not settle which clock the panel is on, and this page does not guess. The cause of the three-hour difference is not established here.

So the documents do not establish that the consultation took place. They establish that it was recommended, approved twice, booked, reminded about, confirmed with the practice, and paid for. A diary entry is evidence of intent, not of attendance, and a receipt whose own timing cannot be placed against the appointment cannot close the gap on its own. Groff’s expectation, written on the morning of the appointment, was that the receipt would reach her after the appointment.¹⁰ Whether what arrived was that receipt is not something any document read for this page says.

The money, and the form of his name that carries it

Two card records carry the charge, and neither of them contains any form of the name “Michael Reed”. Both print the practice’s merchant descriptor, which is “ML Reed MD” in one and the same string in capitals in the other.

A compilation of card statements carries a transaction dated 30 December 2014 against that descriptor, in New York, in the amount of $725.00. The sheet names its account holder in its own running head, “JEFFREY E EPSTEIN”, and gives the statement’s own page as 18 of 60.⁵ A household expense ledger carries a row reading, in its Name column, “ML Reed MD”, in its Memo column the word “medical” followed by a bar, and in its Amount column 725.00.⁶

The date on the ledger row is 12 January 2015, and it is a different kind of date from the one on the statement. The statement’s is a transaction date; the ledger’s is the date the household’s own books record the expense. The ledger’s Date column is month-first, which the page states because it decides the reading: the column runs 01/09, 01/09, 01/09, 01/12, 01/12, 01/30, 03/12, 04/14, 05/14, 06/17, 06/17, 07/15, 07/28, 08/11, 08/28, 09/29, 11/03, 12/22, and the entries 01/30, 04/14, 09/29 and 12/22 can be read no other way.⁶

The same ledger page books charges to Thomas J. Magnani D.D.S. and to the salon business of Frederic Fekkai, one of the latter on the same date as the Reed row.⁶ The card statement carries a salon charge of its own on 30 December 2014, which is the day of the appointment.⁵ The 725.00 is the only amount this page rests any claim on, and that figure was read at the image on both records. The two agree to the cent. The one other figure taken from either record is the salon charge given at ⁵, which is reproduced only to date the salon traffic of the same day and carries no claim about Reed.

What the documents read for this page do not show

• No document read for this page was written, sent, signed or received by Reed. He is not a sender, a recipient, an addressee or a signatory anywhere in the nineteen. His practice is twice a payee, and two messages record contact about the appointment, in each case by somebody else.
• Nothing records what the consultation was for. The word “medical” on the ledger row is followed by a bar, and nothing elsewhere fills it in.
• Nothing records any later contact. The record for this subject stops with the January 2015 ledger row.
• Nothing records him knowing who was paying. No document read for this page shows Epstein’s name reaching the practice, and the one message that mentions payment to the practice describes a telephone call made by the office.
• No statement by Reed about any of this was found. A search outside the released files was made for this page and is described in the next section; it found no public response by him, and this page does not treat that as evidence that he has made none.

Outside the released files

One web search and four pages fetched and read on 19 October 2026; this is not a survey. Two directory biographies describe a Michael Reed practising dermatology at New York University with a surgical specialism in hair transplantation, which matches the affiliation and the field the 2014 message gives him.¹⁸ Neither is a statement by him and neither mentions these documents or Epstein. The web address printed on two of the 2014 sheets does not resolve today, which was checked directly and against a control address that does resolve.¹⁸

No public statement by Michael Reed about these documents, about Epstein, or about this appointment was found. That is the result of one search and is reported as an empty search, not as a finding that no such statement exists. He has not been asked about any of this by anyone, so far as that search reaches, and the absence of a response here is an absence of a question.

How the record was searched, and what this page has not read

The unit in every figure below is DOCUMENTS, meaning distinct document identifiers, not sheets and not occurrences. The vocabulary is given in full so that any figure here can be re-computed.

Seven name arms were run, each separately: "Michael Reed" reaching 14 documents, "M Reed" 12, "Dr Reed" 7, "Reed Michael" 3, "Reed MD" 2, reedmd 2 and mreed 9. Those seven arms reach 47 documents between them, and 17 of the 47 concern this subject. A negative control, Reedzzz, returns 0.

The other 30 are nine other things, and they were read rather than counted. They partition exactly, and the nine counts sum to 30:

what it is	documents
A correspondent of Epstein’s who shares the surname and the middle initial, in LinkedIn digests and in an exchange about another man’s address book	8
A routing code, mreed, printed on Amazon packing slips	8
A psychologist of the same surname and a different forename, retained as an expert in litigation against Epstein, together with a deposition index and a case to-do list naming him	3
Damage in the released text, inside heavily garbled passages where no person is named at all	3
A dating-site username, ReeDMD, in two circular emails	2
A bank’s Caribbean postal address, in which “Road Town” is misread as a personal name	2
A news clipping in which the word “agreed” is misread as the surname	2
A court reporter’s certificate, signed by a different person of the same surname	1
An Outlook delivery-tracking block in which the word “Read” is misread as the surname	1
total	30

Six of those nine classes are not a person of that surname at all, and together they account for 18 of the 30. In this corpus the string is produced about as often by the extractor and by machine furniture as by anybody called Reed, which is why the bare surname arm below is sized and not characterised.

Two of this page’s nineteen documents contain no form of his name at all and are reachable by no name query. Both were found by searching the descriptor the record uses for him rather than his name: "hair specialist" reaches 19 documents, of which 8 concern this subject, and "takes no insurance" reaches 3. One of the two is Epstein’s first approval; the other is a second production of his second approval. A page built on name arms alone would have reached neither, and what they add is one of the two acts this page turns on: the first of Epstein’s two approvals. They do not add the earliest dated item here, which is the recommendation of 16 December and is reachable on the plainest name arm of all.

The bare surname Reed reaches 2,476 documents, and this page has not read them. That arm is a common English word and a common surname, its largest rows are aggregate compilations in which whole-sheet co-presence means nothing, and no attempt is made here to characterise it. It is a named, sized, unread remainder and it is not rounded away. The 47 documents on the seven specific name arms were screened for aggregate rows before any count was taken, by grouping every sheet of the 47 by document and printed page and ranking them on character count. The largest single row is 9,017 characters, and the twenty rows above 8,245 characters all belong to just two documents: the card-statement compilation, whose long rows are its own printed payment boilerplate, and a second statement run whose long rows are bank-reconciliation instructions printed on every sheet. Setting those two documents aside, the largest row is 8,245 characters and is an ordinary sheet of an agreement. Neither kind is an aggregate row in which whole-sheet co-presence would mean nothing, and no aggregate row appears on any of the seven arms.

All nineteen were read in the released text, and all nineteen were read at the page image at 400 dpi. Twelve were read at the image when this page was written. The remaining seven were rested on the released text alone until a later pass rendered and read them at 400 dpi: EFTA02087676, EFTA02088109, EFTA02087207, EFTA02087731, EFTA02088050, EFTA02086613 and EFTA02086367. Each footnote below states which reading it rests on. Every passage quoted on this page was read at the image on the sheet it is cited to. The count of nineteen is DOCUMENTS, meaning distinct document identifiers, and it can be re-computed from the provenance line of every footnote below.

Footnotes

1. Message of 16 December 2014 to Jeffrey Epstein, subject line “Hair Specialist for ▮ from Dr. Merrell”, one sheet, EFTA02515367 p. 1. Read at the image at 400 dpi. The sender is covered by the producing party’s bar and the name in the subject line is covered; nothing is supplied for either. The produced header Sent: Tuesday, December 16, 2014 11:06 PM carries no zone: read as coordinated universal time and converted to New York (EST, five hours behind), 23:06 − 5:00 = 18:06 on the same date, 16 December 2014. Sign control: 18:06 + 5:00 = 23:06, the stamp as printed. The sheet also prints a machine field, date-received 1418771183, a unix epoch: 1418771183 is 23:06:23 on 16 December 2014 in universal time, which is 18:06:23 on the New York clock, so the unzoned header is universal time and the conversion agrees to the minute. Sign control: 18:06:23 + 5:00 = 23:06:23, the epoch as printed. The clause after the quoted words calls him a hair specialist; its apostrophe is destroyed by the production’s quoted-printable transfer encoding and the damaged token is not reproduced. A web address built from his name follows, run together with property-list furniture by the same encoding. A pin checker reports this quoted subject line as unresolved, and for this one string that is expected rather than a defect: the quotation carries a producing party’s bar, the extraction renders that bar as nothing, and the two strings therefore cannot match. The quotation was read at the image. The subject line is printed twice on this page, in the body above and in this footnote, and the number of rows a checker prints for it is not that number: a checker may test a footnote’s own quotations against every line that cites the footnote, so it can raise rows against lines that do not carry the string and none against the body line that does. Nothing else on this page is excused by this note. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

2. Lesley Groff to the person the appointment was for, 18 December 2014, EFTA02087676 p. 1; a second copy carrying the same message inside a longer thread is EFTA02088109 p. 1. Both read at the image at 400 dpi. The produced header given here is the first sheet’s; the second sheet’s own covering header is later and nothing rests on it. Sent: Thur 12/18/2014 3:15:16 PM carries no zone: 15:15:16 − 5:00 = 10:15:16 on the same date. Sign control: 10:15:16 + 5:00 = 15:15:16. That is five minutes and twenty-four seconds after the approval at ⁹, on the same converted clock. The recipient is not named here. The ellipses inside the quoted passage are the writer’s own; the leading ellipsis is this page’s, and marks the opening greeting, which this page does not reproduce. ↩ ↩² ↩³

3. Lesley Groff to Epstein, 18 December 2014, EFTA01747949 p. 1. Read at the image at 400 dpi. The produced header Sent: Thursday, December 18, 2014 7:04:26 PM carries no zone: 19:04:26 − 5:00 = 14:04:26 on the same date. Sign control: 14:04:26 + 5:00 = 19:04:26. The fee is printed $725.00 at the image; one other production renders the dollar sign as the letter S, which is a known extraction fault and not the sheet. ↩ ↩² ↩³ ↩⁴

4. Lesley Groff to the person the appointment was for, copy to Bella Klein, 30 December 2014, subject line “Dr. Michael Reed today at 2:30”, EFTA02087731 p. 1. Read at the image at 400 dpi. The produced header Sent: Tue 12/30/2014 3:02:53 PM carries no zone: 15:02:53 − 5:00 = 10:02:53 on the same date. Sign control: 10:02:53 + 5:00 = 15:02:53. That is one minute and eight seconds after the message at ¹⁰. The greeting is covered on the sheet, and the leading ellipsis in the quoted passage is this page’s and marks it. ↩ ↩²

5. Card-statement compilation, EFTA01671962 p. 7007, carrying a transaction dated 12/30/14 against a merchant descriptor printing the practice’s name in capitals, in New York, at $725.00. Read at the image at 400 dpi. The object is a 9,824-sheet compilation; two attempts to render this sheet timed out before a third completed, which is recorded because the sheet is one of the two places the money appears. The sheet’s running head names the account holder, JEFFREY E EPSTEIN, and the same running head numbers the sheet page 18 of 60 of that statement, so the compilation page and the statement page are different numbers; the foot of the sheet carries Bates stamps and not the statement’s own page number. The sheet’s own printed Bates stamp differs from the identifier above by exactly the page offset, which is a property of this production generally and not of this document. No account ending, merchant identifier or telephone number is reproduced from this sheet. A salon charge of $94.05 stands two rows above on the same date. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

6. Household expense ledger, EFTA01125851 p. 2. Read at the image at 400 dpi. The row reads, across its columns, 01/12/2015, ML Reed MD, medical followed by a bar, and 725.00. The account name at the head of the table and the beneficiary names in several Memo cells are covered. The Date column is month-first, established from the column’s own sequence as set out in the body above; the row therefore books 12 January 2015. That is a ledger booking date and not a service date, and the transaction date for the same charge is the 30 December 2014 at ⁵. The table’s total is not reproduced here because no claim on this page rests on it. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

7. Thread “Appt w/Hair Specialist”, 18 December 2014, EFTA02599407 p. 1, read at the image at 400 dpi. Lesley Groff to Epstein. The produced header Sent: Thursday, December 18, 2014 2:14 PM carries no zone: 14:14 − 5:00 = 09:14 on the same date. Sign control: 09:14 + 5:00 = 14:14. The sheet also prints a machine field, date-received 1418912019, a unix epoch: 1418912019 is 14:13:39 on 18 December 2014 in universal time, which is 09:13:39 on the New York clock, agreeing with the converted header to the minute. Sign control: 09:13:39 + 5:00 = 14:13:39, the epoch as printed. This sheet prints the recommending doctor’s surname as =errell, the initial letter destroyed by the production’s transfer encoding, and the quotation above reproduces that damaged form rather than silently mending it, because a page quotes the sheet it cites. The name of the person the appointment concerns is covered. The sheet’s tail runs into property-list furniture. ↩ ↩² ↩³

8. Thread of 30 December 2014, EFTA02088050 p. 1; the same exchange is quoted again lower down a longer message sent later the same evening, EFTA02086613 p. 1, which is a different message and not a second production of this one. Both read at the image at 400 dpi. Groff’s covering message is produced-stamped Sent: Tue 12/30/2014 4:01:04 PM, which carries no zone: 16:01:04 − 5:00 = 11:01:04 on the same date. Sign control: 11:01:04 + 5:00 = 16:01:04. The quoted reply below it is timed On Dec 30, 2014, at 10:58 AM by the quoting software and carries no offset; the page does not convert it and does not rely on it for anything but the order, which the thread itself fixes. ↩ ↩²

9. Epstein’s reply, one word, “Yes”, EFTA02087539 p. 1, read at the image at 400 dpi. The produced header Sent: Thur 12/18/2014 3:09:52 PM carries no zone: 15:09:52 − 5:00 = 10:09:52 on the same date. Sign control: 10:09:52 + 5:00 = 15:09:52. This is one of the two same-message twins establishing that the produced header is universal time: EFTA02087620 p. 1 and EFTA01747949 p. 1 both quote this message as sent “at 10:09 AM”, which is the converted reading to the second. ↩ ↩² ↩³

10. Forwarded chain of 30 December 2014, Lesley Groff to Bella Klein, EFTA02087620 p. 1. Read at the image at 400 dpi. The covering message’s produced header Sent: Tue 12/30/2014 3:01:45 PM carries no zone: 15:01:45 − 5:00 = 10:01:45 on the same date. Sign control: 10:01:45 + 5:00 = 15:01:45. The forwarded message from Epstein carries an explicit zone, Date: December 18, 2014 at 3:03:06 PM EST, and needs no conversion. The four stops in “amex….” are the writer’s own; the trailing ellipsis is this page’s and marks an elision made here. This sheet quotes Groff’s message of ³ as sent at 3:04 PM, which is one hour later than the converted reading of its own produced stamp and one minute after the zoned reply it provoked; the page relies on the converted reading and records the discrepancy rather than explaining it. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

11. A second production of the same reply, EFTA02087207 p. 1. Read at the image at 400 dpi. At the image the produced header prints whole, Sent: Thur 12/18/2014 8:03:06 PM; it is the extraction of this sheet, not the sheet, that renders it as Sent: Thu followed by a damaged date and 8:03:06 PM. The header carries no zone: 20:03:06 − 5:00 = 15:03:06, which is 3:03:06 PM, the zoned reading at ¹⁰ to the second. This is the second and stronger of the two same-message twins, because the two readings of this one message differ in whether a zone is stated at all, which is what makes the conversion checkable rather than assumed. This document contains no form of the subject’s name and is reachable by no name query; it was found on the descriptor arm "takes no insurance". ↩ ↩²

12. Google Calendar notification, 30 December 2014, EFTA02087459 p. 1. Read at the image at 400 dpi. The produced header Sent: Tue 12/30/2014 2:49:48 PM carries no zone: 14:49:48 − 5:00 = 09:49:48 on the same date. The event’s own time carries a zone on the sheet, which prints When Tue Dec 30, 2014 10am – 10:30am Eastern Time. The interval between the two is therefore zoned against converted: 10:00:00 − 09:49:48 = 10 minutes 12 seconds. Sign control: 09:49:48 − 10:00:00 = minus 10 minutes 12 seconds, so the notification precedes the event, which is what a reminder does. The Calendar and Who fields are covered, as is a telephone number inside the title. ↩ ↩²

13. Jeffrey Epstein to Lesley Groff, one sheet, EFTA01732112 p. 1. Read at the image at 400 dpi. His whole message is “yes”. The produced header Sent: Wednesday, December 17, 2014 1:12 AM carries no zone and stands inside the rollover window: 01:12 − 5:00 = 20:12 on 16 December 2014, the previous day. Sign control: 20:12 + 5:00 = 01:12 on 17 December, the stamp as printed. The sheet also prints a machine field, date-received 1418778701, a unix epoch: 1418778701 is 01:11:41 on 17 December 2014 in universal time, which is 20:11:41 on 16 December on the New York clock, so the machine field puts this message on the evening of the 16th independently of any inference about the header’s clock, and agrees with the conversion to the minute. Sign control: 20:11:41 + 5:00 = 01:11:41 on 17 December, the epoch as printed. This sheet quotes the preceding message as sent at 7:06 PM on 16 December, where the converted reading of its own produced stamp is 6:06 PM; the one-hour difference is recorded and not explained. This document contains no form of the subject’s name and is reachable by no name query; it was found on the descriptor arm "hair specialist". ↩ ↩² ↩³

14. Lesley Groff to the person the appointment was for, copies to Jeffrey Epstein and Bella Klein, 18 December 2014, subject line “Dr. Michael Reed”, EFTA01747373 p. 1. Read at the image at 400 dpi. The produced header Sent: Thursday, December 18, 2014 9:19 PM carries no zone: 21:19 − 5:00 = 16:19 on the same date. Sign control: 16:19 + 5:00 = 21:19. The sender, the recipient and the greeting are covered. The leading ellipsis in the quoted passage is this page’s, and marks that covered greeting. =ues and =f reproduce the sheet, which prints an equals sign in ink in both places where the production’s transfer encoding has replaced the first character of a word; they are reproduced as printed rather than repaired, and no bar is drawn for them because the producing party covered nothing there. The practice’s street address and telephone number are covered on the sheet and are not supplied here, in keeping with this site’s standing practice on addresses and telephone numbers for everyone. ↩

15. Reply of 18 December 2014 to Lesley Groff, copies to Jeffrey Epstein and Bella Klein, EFTA01747548 p. 1. Read at the image at 400 dpi. The produced header Sent: Thursday, December 18, 2014 10:16:28 PM carries no zone: 22:16:28 − 5:00 = 17:16:28 on the same date. Sign control: 17:16:28 + 5:00 = 22:16:28. ↩

16. Message of 30 December 2014 to Lesley Groff, EFTA02086367 p. 1. Read at the image at 400 dpi. The produced header Sent: Tue 12/30/2014 8:54:50 PM carries no zone: 20:54:50 − 5:00 = 15:54:50 on the same date. Sign control: 15:54:50 + 5:00 = 20:54:50. The sheet reproduces the 30 December exchange quoted above and adds nothing further about Reed. ↩

17. Lesley Groff to Bella Klein, 31 December 2014, subject line ”▮ Dr. Reed Visit receipt”, EFTA00355002 p. 1. Read at the image at 400 dpi. The covering header carries an explicit zone, Date: Wed, 31 Dec 2014 15:30:54 +0000: converted to New York, 15:30:54 − 5:00 = 10:30:54 on the same date. Sign control: 10:30:54 + 5:00 = 15:30:54. The forwarded notification’s own header also carries a zone, Date: December 30, 2014 at 4:00:53 PM EST. The panel inside it states a receipt time of 01:00 PM with no zone given for that field, and the page draws no conclusion from the three-hour difference. The fax’s sender field inside the panel is covered by the producing party’s bar and nothing is supplied for it. No contact value of any kind is reproduced from this sheet. The attachment named in the header is not in the release as far as this page’s searches reach. ↩

18. Outside the released files, 19 October 2026. One web search was run; four pages were fetched directly and read. A WebMD directory biography, fetched at HTTP 200, describes “Michael Reed, MD” as “an assistant professor of clinical dermatology at New York University Medical Center” maintaining “an active private practice of general and cosmetic dermatology in Manhattan”, whose “areas of special surgical expertise include hair transplantation and liposculpture”, and who “is an attending physician in charge of microsurgical hair transplantation at NYU”. A user-edited wiki page, fetched at HTTP 200, gives his name as “Michael Lorin Reed” and describes him as a clinical associate professor in New York University’s dermatology department; it is user-edited and is not relied on here for anything the WebMD entry does not also say. Neither page mentions Epstein or these documents, and neither is a statement by him. The web address printed on the 2014 sheets returns no DNS record today, checked three times across two hostnames, with a control hostname that resolves normally. No public statement by him about this material was found, and that is an empty search rather than a finding of absence. ↩ ↩²