Category: ghost

Ghost Expedition Worcester County, Snow Hill Maryland:…

Ghost Expedition Worcester County, Snow Hill Maryland: Nassawango Furnace Archaeological Site/Furnace Town Living Heritage Village

According to the historian Mercedes Quesada-Embid, the story of Furnacetown and the Nassawango Iron Furnace in Snow Hill Maryland involves transitions through eras of colonial expansion, industrial boom and bust, abandonment followed by environmental renewal, conservation and historic preservation 

The wetlands, forests and coasts in the greater Nassawango Creek were initially settled by the Pokemoke and Assateague tribes. In the Pokemoke language, Nassawango means “the ground between the streams.”  The Nassawango Creek is the main tributary of the Pokemoke river which empties into the Chesapeake Bay

The initial contact with the native tribes was made by the Englishman, Captain John Smith in 1608.  Native villages were not concentrated nor permanent settlements; families and tribes relocated as seasons changed.  Small areas and passages were cleared for hunting and gathering, farming and protection

The initial exchanges between natives and settlers were cooperative during a short-lived fur trading industry.  As settlements expanded and colonial land uses turned toward tobacco farming, forest areas were cleared for agricultural use.  Deforestation and the rise of large-scale plantations altered the natural habitat  accelerating the disappearance of the native way of life, leading to conflicts over land  

Native tribes were relocated to a portion of the Nassawango near present day Snow Hill known as Askiminkonson, which in Algonquin means “stony place where they pick early berries.” The swampy lowland was considered not suitable for farming.  Native petitions to the English government were unsuccessful and the reservation dwindled.  By 1750 there there was no native presence


Nassawango creek entered an industrial era when a charter was granted to the Maryland Iron Company in 1829.  The company claimed 5000 acres of forest and swamp lands, which were rich in bog ore, as well as a gristmill, sawmill and millpond. A hot blast furnace was built by 1831. From oral histories, the company erected a

“furnace town”

with streets, company stores, a blacksmith, a dressmaker, hotel, post office and church

The company ran into financial troubles by 1832 and was fully acquired by a wealthy Philadelphia-based industrialist named Ben Jones by 1834.  In 1835, the operation was leased to a Thomas Spence, a young lawyer based in Snow Hill.  For a time the operation flourished and was producing 700 tons of pig iron annually. The furnace ran for 24 hours per day for 32 weeks out of the year.  Light from the orange flames produced by the furnace could be seen for miles

However, poor ore quality and declining market demand led to closure by 1850. Nearly all of Furnace Town’s residents departed leaving a ghost town. After the furnace closed, the property was used by successive owners for timber rights.  However, for the next 100 years, the land was untouched and ecological processes began to restore the wetlands, forests and habitats that had been cleared or polluted by slag by industrialization  

Land surrounding the old furnace was donated to the Worcester County Historical Society (WCHS)

in 1962. The Nature Conservancy (TNC) acquired adjoining lands from 1977 to 1981.  WCHS and TNC partnered to preserve area history and ecology and beginning in 1977 had moved several historic buildings to site to form the Furnace Town Living Heritage Museum.

The partnership formed the Furnace Town Foundation in 1982.   

The Pokemoke Forest has many folklore and legends to include the Hook-Man, Goat-Man, fireballs and lights, slave and swamp ghosts, and elemental creatures.  Furnace Town itself is said to haunted by the ghosts of several former area residents to include the late Sampson Harmon, the town’s last resident

Sampson Harmon was a free African-American born in Nassawango Hills. He was said to be a

“big, tall, fast, and strong man.”  He was the “go-to” worker at the iron furnace and worked very hard to provide for his family. Sampson always wore a hat and was fictionalized as “Sampson Hat” in George Alfred Townsend’s novel “The Entailed Hat”


When the iron furnace closed Sampson insisted on staying in Furnace Town. His dying wish was to have his ashes left at his homestead but this was not granted. His ghost is said to wander and guard the area

The ghost expedition will seek to obtain “drop-in” communications connected with the rich and storied history of the Nassawango


‘Folk Tale Trilogy’ Is Celebration Of Stories. (1988, Jul 6). The Daily Times (Salisbury MD).

Furnace Town Living Heritage Village. (2018). Nassawango Furnace Archaeological Site, Worcester County, Snow Hill MD. Furnace Town.

Kester-McCabe, D. Tales of Snow Hill. Delmarva Almanac.

LeVan, K., and Reiten A, (2006). The Snow Hill Historic District. Snow Hill Historic District Commission. Town of Snow Hill Maryland.

Lutz, L. (2005, Jun 1). Nassawango’s furnace – and forest – rising from the ruins. Bay Journal.

Miller, N. (1973, Apr). National Register of Historic Places Registration. Nassawango Iron Furnace Site. Maryland Historical Trust.

Quesada-Embid, M. (2004). Five Hundred Years on Five Thousand Acres: Human Attitudes and Land Use at Nassawango Creek, Native Americans of the Delmarva Peninsula. Edward H. Nab Research Center for Delmarva History and Culture, Salisbury University Libraries, Maryland Shared Open Access Repository (SOAR).

Robbins, M.W. (1972). The Maryland Iron Industry. Manuscript prepared for the Maryland Bi-centennial Commission, Annapolis, Maryland.

Runkle, S. A. (2003, Sep). Native American Waterbody and Place Names within the Susquehanna River Basin and Surrounding Subbasins Publication 229. Susquehanna River Basin Commission.

Sampson Harmon: Furnace Town’s Resident Cat Collecting Ghost. (2012, Oct 30). ShoreBread.

Searching for history at Furnace Town. (1990, Aug 5). The Daily Times (Salisbury MD).

Teich, I. 14 Myths and Legends Surrounding Maryland’s Haunted Pocomoke Forest. Ranker.

Touart, P. (2009). Worcesters’s African American Heritage. Worcester County Tourism.

Worcester County, Maryland: Historical Chronology. Maryland State Archives.


Bourne, M.O., photographer. (1969, Nov). Furnace Stack, looking southeast. Nassawango Furnace Archaeological Site, Worcester County, Snow Hill Maryland. Maryland Historical Trust.

Bourne, M.O., photographer.

(1969, Nov). Detail, hot air apparatus, looking northeast. Nassawango Furnace Archaeological Site, Worcester County, Snow Hill Maryland. Maryland Historical Trust.

DETAIL, ¾ VIEW OF HOT BLAST STOVE ON TOP OF FURNACE SHOWING CAST-IRON RETORTS AND TURNED HEAD (WHERE RAW MATERIALS WERE LOADED INTO FURNACE). Nassawango Iron Furnace, Furnace Road, 1.2 miles west of Maryland Route 12, Snow Hill, Worcester County, MD. Historic American Engineering Record, Library of Congress.

HAER MD,24-SNOHI.V,2- (sheet 6 of 12) – Nassawango Iron Furnace, Furnace Road, 1.2 miles west of Maryland Route 12, Snow Hill, Worcester County, MD. 

Historic American Engineering Record, Library of Congress.

Sculpture of Sampson Harmon. (2018). Furnace Town Living Heritage Village. Furnace Town Foundation.

Photograph of Sampson Harmon. (2009). In Worcesters’s African American Heritage. Worcester County Tourism. Courtesy of the Julia A Purnell Museum.

Robbins, M.W., photographer.

(1972). Furnace casting hearth, looking west. Nassawango Furnace Archaeological Site, Worcester County, Snow Hill Maryland. Maryland Historical Trust.

Nassawango Iron Furnace, looking southeast. (2018). Worcester County, Snow Hill Maryland. Furnace Town Living Heritage Village.

A woman in Alabama has claimed to see God walk…

A woman in Alabama has claimed to see God walking in the clouds. (Via: YouTube)

Little Girl’s ‘imaginary friend’ definitely…

Little Girl’s ‘imaginary friend’ definitely sounds like a ghost. What do you think ?

Cafe closed after Paranormal Activity – Video – Unexplained Mysteries

Cafe closed after Paranormal Activity – Video – Unexplained Mysteries




Moving Milk Bottle Cap Mysteries: Ghosts, Poltergeists, or the…

Moving Milk Bottle Cap Mysteries: Ghosts, Poltergeists, or the Ideal Gas Law?

Videos of moving bottle caps on milk containers appearing across social media have captured degrees of media interest

In some news publications, the phenomena had been sourced as having a paranormal origin, albeit a bit tongue in cheek. Other posts have also attributed the mystery movements to ghosts, poltergeists or haunted phenomena 

However most posts, like the one featured, involve persons who are fascinated by the inexplicable movements in themselves. The milk containers in the videos are often nearly empty.  This observation helps in pointing toward a normal versus paranormal explanation for the events at hand

The Ideal Gas Law conveys that the state of an amount of gas is determined by its pressure (P), volume (V), and temperature (T).  The law is usually expressed in the equation:

                                                      PV = nRT

Where n is the number of moles (molecular mass), R is a the gas constant

When the milk container is removed from the refrigerator, the temperature (T) rises as the bottle warms up. The volume (V) of the container is unchanged but there is greater kinetic energy among the (n) gas molecules within it.  The result is higher pressure (P) within the container

The Ideal gas Law equation requires adjustments for high pressures and temperatures, and there should be no other attractive forces among the molecules. But it is a good approximation for the conditions seen in the videos 

Another example of the Ideal Gas Law in action can be seen in the jumping coin experiment


Crashcourse. (2013, May 7). The Ideal Gas Law: Crash Course Chemistry #12. YouTube

Delaney76. (2008, April 20). Moving Milk Bottle Cap. YouTube

Elearnin. (2013, May 7). Coin jump up trick revealed | Science experiment. YouTube

Keegan, N. (2015, Dec 24). Spooky moment ‘ghost’ lifts lid off bottle of semi-skimmed milk. The Sun (2015, Dec 22). Watch terrifying moment ‘ghost’ lifts lid off bottle of semi-skimmed milk. The Mirror

Nave, C.R. (2017). Hyperphysics: Ideal Gas Law. Department of Physics and Astronomy. Georgia State University

Wikipedia. (2018, May 11). Ideal gas law

Exploring Nursing Ghost Stories through Machine Learning: Topic…

Exploring Nursing Ghost Stories through Machine Learning: Topic Discovery with Latent Dirichlet Allocation

NOTE: Click to open graphics for an expanded and clearer view of the findings they contain  

As reported in earlier posts, the web site hosts a long-running moderated discussion thread called “Nursing Ghost Stories” (NGS).  The NGS collection spans over a decade (2005-2017) amounting to 199 pages as of the time of this writing.  As a dataset NGS contains multiple first and second hand accounts and commentary on paranormal type experiences 

The archive contains classic examples of hauntings and poltergeist phenomena.  Patients were generally the percipients in ghost experiences.   Sometimes the ghosts in question appeared to be former nurses in period dress, or former doctors and patients, or former area residents.  However, these kinds of paranormal experiences did not dominate the collection  

In actuality, the NGS archive conveys several varieties of psi and post-mortem survival phenomena.  The archive contains several examples of extrasensory perception and presentiment in particular  

There were also examples of after-death communication (ADC), which are sensed-presence or apparitional experiences involving deceased family members or friends.  Unlike hauntings which are place-centered, ADC encounters are person-centered involving meaningful coincidences (or synchronicities) for the percipients

The archive contains several reports of near-death experiences (NDEs). However, the more representative encounters involved nearing death awareness (NDA) type experiences.  In NDA situations, terminally-ill patients experiencing death-bed visions will have perceptions of welcoming apparitions of deceased relatives or loved ones

  • Terminal patients will also appear to hold conversations with persons who are not physically present in their room.  Sometimes nurses described these aspects of NDA experiences as dementia
  • It is also not uncommon for gravely-ill patients to be alert and conversant in their final hours before death, a phenomenon called “terminal lucidity”

Provided below are examples of exchanges regarding NDA situations as characterized by nurses working in long-term care and palliative care settings 

I’ve been a hospice nurse for 5 years. I have been with hundreds of people at the time of their death & I can tell you first hand that if the patient is alert enough to speak, you’ll hear them talking to loved ones that have already passed over

That is so true. I, too am a hospice nurse and when pts. start talking to their dead relatives, you know that they have about a week MAX before they are gone

From experience I’ve learned that when a pt tells you they’re going to die…they usually do…and if they start talking to dead family members…they usually die…it’s like the family members have come to take them…..

As a follow-on to the earlier wordcloud project, we wondered whether unsupervised machine learning, specifically topic generation models, could discover the abovementioned themes in the NGS archive 

  • Generative topic models

    view documents as having a latent semantic structure of topics that can be inferred from co-occurrences of words in documents  

  • For this project, the Latent Dirichlet Allocation (LDA) topic model was employed.  LDA views documents as probability distributions over topics and topics as probability distributions over words
  • All documents share the same collection of topics, but each document contains those topics in different proportions.  The LDA algorithm samples words across topics until it arrives at topics and word selections that most likely generated the documents

Various packages and libraries for natural language processing within Python were used to include: the Natural Language ToolKit (NLTK) for processing the data set; scikit-learn to prepare and fit the LDA model; pyLDAvis to display the results and t-Distributed Stochastic Neighbor Embedding (t-SNE) to map topic distances

The project pipeline involved: data set processing; conversion of words and documents into a document-term matrix and vector space; fitting the LDA models; and displaying the results

Processing. The data set was decomposed into 199 documents from its constituent web pages.  In contrast to the wordcloud project, the set of stopwords was enlarged to find meaningful insights in the NGS archive

  • The core set of stopwords consisted of commonly-used prepositions, conjunctions, and contractions.  Stopwords from the wordcloud application were used as a start point for this purpose
  • Since the archive consisted of first or second hand accounts, words related to stories and/or storytelling were added to stopwords, along with words related to the maintenance of the thread
  • Since spontaneous experiences can occur at any moment, words conveying times were removed.  While many experiences were singular events, numeric references involving ordinal (e.g. one, two) and cardinal (e.g. first, second) rankings were removed
  • Titles of persons were removed (e.g. Mr., Mrs., etc.); however, person and gender types (e.g. man, woman, etc.) and interpersonal relationships (e.g. family, friends, or strangers) were preserved
  • Domain-related words relating to patient care or standard procedures were removed (e.g. hospital, unit, shift, staff, work, station, monitor, code)

Conversion. Vector transformations converted the data set into a document-term matrix for mathematical processing.  The rows of the matrix correspond to documents with columns corresponding to the frequency of a term

  • Count vectorizers count word frequencies.  Term Frequency-Inverse Document Frequency (TF-IDF) vectorizers normalize (divide) word counts by their frequency in the documents
  • Both vectorizers converted words to lower case and removed non-word expressions. The vectorizers were parameterized to look for bigrams (or words that were often used together) 

Model Fit/Display. The LDA model was fitted using ten topics.  Words within topics were sorted and ranked with respect to their frequency in and relevance within a topic

  • The LDA model was fitted with using Count and TF-IDF vectorization and ran with a maximum of 100 iterations.  LDA model results were displayed using pyLDAvis and t-SNE to map topic distances

Results. Although topics produced from the model are unlabeled, words within topics usually can be woven into a coherent theme

The first four pyLDAvis graphs provide the top 30 words and bigrams in Topics 1 through 4 using Count vectorization  

  • Topic 1 is the most representative of the body of stories in the thread and generated around 86% of the content.  Words in Topic 1 included: “nurse” and “patient”; both nurses and patients were percipients and sometimes sources of “ghost” experiences.  If apparitions represented unrecognized persons, patients had “asked” whom they “saw.”  Many apparitional encounters involved patients who were “heard” “talking” to deceased “family” members or a “friend.“  These telepathic types of apparitions were often described as “sitting” near the bedsides of patients, or transiting their rooms or into an adjacent “hall” on their “floor.”  Overall, this could be considered an apparitional experiences topic

  • Topic 2 is derived from user commentary and seems reflective of internal varieties of psi functioning. Words in Topic 2 included:  “dreams”, “feel(ings)” and a “sense” of awareness or presentiment of events that were happening or about to “happen”, usually in connection with the deaths of family members. In other cases the dreams were possible telepathic connections with lost “loved” ones. Overall, this can be considered a extrasensory perception topic and it generated 7% of the content  
  • Topic 3 appears reflective of external forms of psi and survival phenomena to include auditory and physical encounters commonly associated with hauntings and poltergeists.  Words in Topic 3 included: “haunted”, “voice(s)”, and other imitative sounds such as “music.”  There were also reported instances of anomalous telephone contact possibly involving “phone” calls from the dead and “strange” behaviors of televisions, call lights and other electrical appliances.  Overall, this could be considered a hauntings and poltergeists topic and it generated around 4% of the content 
  • Topic 4 is also derived from user commentary and seems reflective of general discussions on the paranormal, religious and exceptional experiences.  Discussions included: “paranormal” television, “movie” and “radio” entertainment;  synchronicities (meaningful coincidences) and “photo” and other evidence from paranormal investigations.  Discussions also involved ghost stories outside a nursing context; some were urban legends and a few were probably larks.  Overall, this could be considered a paranormal discussions topic and it generated around 3% of the content

The fifth pyLDAvis graph provides the top 30 words in Topic 1 using TF-IDF vectorization.  

  • The findings were close to those encountered for Topic 1 with the Count Vectorization.  However, it appears to be a combined apparitional experiences and extrasensory perception topic accounting for 94% of the content.

     This consolidation arises from the fact that TF-IDF vectorization lowers the contribution weight of commonly used words

This project again demonstrates the usefulness of topic generation models for finding meaningful patterns in masses of unlabeled or unstructured data.

  The LDA topic discovery method indicated several varieties of psi and survival experiences that went beyond ghost stories 

  • Many apparitional encounters described in the archive represented the intersection of nearing death awareness (involving death-bed visions of welcoming apparitions) and after-death communication experiences (involving apparitions of deceased family members and friends)
  • Even though the algorithm knows nothing intrinsically about the above experiences, the model was able to infer topics and words corresponding to the most representative kinds of encounters 

Greater insights could be gained by structuring the NGS dataset and labeling the experiential elements within it.  Follow-on research could employ semi-supervised methods to train models to classify types of psi and survival experiences and to find correlates within them  

Specifically, deep learning models could be trained on the semantics around typologies of apparitions with tagged documents.  Parapsychology categorizes apparitions along four lines: living agent; crisis; post-mortem; and haunting  

  • If an apparition is seen within ±12 hours of a person’s death, that represents a crisis apparition 
  • If an apparition is seen 24 hours or more after a person’s death, that apparition is post-mortem
  • If the apparition is of a long-deceased person and has a location affinity, that is a haunting apparition

Nonetheless, the apparitional experiences in NGS appear roughly consistent with survey results elsewhere.  Apparitional experiences rarely occur in the general population, but when they do, the apparitions are likely to represent recognized persons, known to the individuals who are perceiving them


Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent dirichlet allocation. Journal of machine Learning research, (Jan), 993-1022.

Gauld, A., & Cornell, A. D. (1979). Poltergeists. Routledge Kegan & Paul.

Kircher, P. and Callanan, M. (2017, Dec 14).  NDEs and Nearing Death Awareness in the Terminally Ill. International Association for Near Death Studies (IANDS).

Natural Language Toolkit: NLTK 3.2.5 documentation. (2017, Sep 24). NLTK Project.

Pearson, P. (2014). Opening Heaven’s Door: What the Dying May be Trying to Tell Us about where They’re Going. Random House Canada. Sponsored

Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., … & Vanderplas, J. (2011). Scikit-learn: Machine learning in Python. Journal of machine learning research, 12(Oct), 2825-2830.

Sievert, C., & Shirley, K. (2014). LDAvis: A method for visualizing and interpreting topics. In Proceedings of the workshop on interactive language learning, visualization, and interfaces (pp. 63-70).

What’s Your Best Nursing Ghost Story? (2017, Oct 30).


pyLDAvis Graph of Topic 1 (Count Vectorization) from Nursing Ghost Stories Corpus. (2018, Apr 08). © Maryland Paranormal Research ®.  All rights reserved.

pyLDAvis Graph of Topic 2 (Count Vectorization) from Nursing Ghost Stories Corpus. (2018, Apr 08). © Maryland Paranormal Research ®.  All rights reserved.

pyLDAvis Graph of Topic 3 (Count Vectorization) from Nursing Ghost Stories Corpus. (2018, Apr 08). © Maryland Paranormal Research ®.  All rights reserved.

pyLDAvis Graph of Topic 4 (Count Vectorization) from Nursing Ghost Stories Corpus. (2018, Apr 08). © Maryland Paranormal Research ®.  All rights reserved.

pyLDAvis Graph of Topic 1 (TF-IDF Vectorization) from Nursing Ghost Stories Corpus. (2018, Apr 08). © Maryland Paranormal Research ®.  All rights reserved.

paranormaldaily: Story found on…


Story found on

 3D Sonogram Reveals Much! Hello. My name is Charity. In March 2011, I lost my father. We had many discussions about the paranormal and I always thought it was crazy. He proceeded to tell me that after he dies, he will appear to me, somehow, someway. Before he died, my son and his wife had their first child. My dad went to the hospital a couple times to see the baby. She was born in February 2011. Not a month later, he passed. Now my son and his wife are expecting again. She recently had to get a 3D sonogram because the baby wasn’t as developed as the doctors would have liked, that and my daughter-in-law has diabetes, which is another reason why they had to do the 3D sonogram.

She was with her mother at the time of the sonogram and didn’t see anything strange or out of the ordinary. They got in their car to leave and her mother wanted to see the printed out pictures of the new baby. As she took a glance at the picture, she noticed there was a man in the picture. After my daughter-in-law started looking at it, she recognized the man in the photo and said it looked like her husband’s grandfather that passed away.

After she had called me and told me about this picture, she sent me a copy on my phone. Sure enough, it looks just like my father. I went to their home and seen the original picture, and it is in 3 or 4 of them! I am now a believer.

I am enclosing one of the sonogram pictures and a picture of my father.